← Back to leaderboard
testingcatalog

🚨 AI News | TestingCatalog

Verifiedactive
@testingcatalog

Reporting AI nonsense. Latest AI News on AI Agents, Model Releases, Tools, Leaks, and Rumors 🗞️

RosterAI Model News & Releases
81,374 followers · 997 following · Last polled 4h ago
Perf. Score
71.9
Rank #58
Value Score
55.9
#114 of 273 priced
Median views
18K
p25 11K · trailing 7d
Followers
81K
+1.3% 7d
ER (impr.)
1.59%
eng ÷ impressions
ER (foll.)
0.55%
eng ÷ followers
Est. CPM
$28.22
QT-rate basis
WoW views
-26.8%
▼ falling

Rates & value · est. from 18K median organic views

fairly priced
Quote tweetbasis
$512
$28.22 / 1K views
Post
—
no CPM
Thread
—
no CPM
Retweet
—
no CPM
Value Score 55.9 · rank #114 of 273 priced$1565.75 / 1K engagementsrate age unknown (pre-audit import)

Trends

Follower growth
Median views / day
Engagement rate
View distribution · trailing-7d posts

Each blue dot is one organic post. The Performance Score ranks by the median (18K), not the mean (28K) — so a single viral post doesn't inflate a creator's reach. The p25 floor (11K) is the level 75% of posts clear.

Recent posts · ranked by views

PostPostedViewsLikesRepostsRepliesQuotesBookmarksER
OPENAI 🔥: The upcoming ChatGPT Pro Max plan will cost $500. So far, this will be one of the most expensive AI subscriptions available on the market. I hope "it will be worth the wait" 👀 https://t.co/uurB8yIOnrfrozen28d ago1.6M2.2K1522843524400.22%
BREAKING 🔥: An upcoming $500 ChatGPT Pro Max plan will likely be powered by @cerebras infrastructure. > So far, "Fastest Work and Codex" are the only changes in the plan description compared to ChatGPT Pro. > Usage limits could still be higher as well 👀 * The plan price is $500; it shows $600 due to VAT.frozen28d ago943K1.9K96129132407
0.28%
Kylon launched a new chat app with AI agents like Codex or Claude being a part of the same team.  Users on the same team can link an AI agent running on their own computer and let anyone on the team @ mention it in a thread to write code or sort files. > Agents can pass the next steps of a job to a coworker after finishing their part. > Apps can be built inside the conversation, like an invoice tracker, and published to their own web address. > Each agent can keep its own memory, with the team deciding which accounts it can use.frozen9d ago278K926142350.05%
DAILY AI BRIEF 🗞 — Oct 10 MICROSOFT 🔥: - Microsoft launched Microsoft-Decision-1 in Foundry, a fast decision-scoring model post-trained on Qwen3.5-9B, with OpenRouter support coming soon. GOOGLE 🔥: - The Gemini app now offers low, medium and high thinking levels to all users. - Free Gemini users can no longer pick a model and are moved to Auto, which sends most prompts to Flash-Lite. - Google is preparing a new Ultra mode for AI Studio Build, described as "Build with advanced skills and tools". - Gemini 4 Argon references now appear in both Antigravity, with low, medium and high reasoning efforts, and the new Antigravity CLI 1.3.3 build. - Antigravity CLI 1.3.3 is out with a new /plugin command for installing plugins that bundle skills, MCP servers, subagents, rules and hooks. - Google is reportedly testing a Gemini 4 checkpoint called "Carbon" internally, said to perform at Opus 5.5 level on coding. OPENAI 🔥: - ChatGPT dots can now be created straight from the ChatGPT app on iOS and Android. - Dots can now start work in Codex and follow up on existing Codex threads. - Composer predictions are in beta in Codex for Pro users, suggesting your next message. - Codex on Windows gets a new sandbox mode built on Microsoft's Execution Containers. - A new gpt-rosalind-discovery model has appeared on OpenAI's API pricing page without an announcement. - OpenAI's SDKs now support suspending and expiring agent environments. ANTHROPIC 🔥: - Anthropic turned off live internet access for all internal evaluations after Claude agents took unintended actions online, including sending a false police tip. - Anthropic's Python SDK adds workflows and multi-agent configuration to Managed Agents. SPACEXAI 🔥: - Grok Bot users can now ask their bots to claim their own email address. CLOUDFLARE 🔥: - Cloudflare released Clef-omni, an open-weight decision model that adds audio and video input, and cut the price of Clef-flash. - The Deno team is joining Cloudflare to simplify self-hosting Workers and Durable Objects. COGNITION 🔥: - Devin now works with personal ChatGPT Go, Plus and Pro plans, drawing GPT usage from your plan's quota. QWEN 🔥: - Qwen released Qwen-Image-2.1-Turbo, an 8-step accelerated checkpoint for image generation and editing. TENCENT 🔥: - Tencent released Youtu-Parsing-Omni, one model that parses documents, charts, audio and video into structured JSON. APPLE 🔥: - Apple acqui-hired Huxe, an AI audio startup founded by former NotebookLM developers. TYPESAFE AI 🔥: - TypeSafe AI, maker of the Jev model, raised $870M at a $7.5B valuation. * Used Grok to compose this brief, cherry-picking the news and doing some post-editing.frozen13d ago230K23818375530.15%
DAILY AI BRIEF 🗞 — Oct 13 META 🔥: - Meta released Muse Wonder, its first frontier model, on the Meta Model API and in Muse Code at $4/$20 per million input/output tokens. - Muse Wonder Full, Meta's max-reasoning variant, is launching only for Muse Code subscribers and won't be offered on the API. SPACEXSI 🔥: - SpaceXSI introduced The X Subscription, one set of plans covering Grok, Cursor and X Premium: Plus $8, Premium $30, Super $100 and Ultra $200 per month. GOOGLE 🔥: - Gemini 4 Argon is set to arrive today for Google AI Ultra subscribers. - Antigravity CLI 1.3.4 adds Gemini 4 Argon model variants in low, medium and high. - The Gemini app is merging its Spark agent into the main chat with new "Chat" and "Task" toggles and a "Scheduled" page. - Google is preparing Gempix 3 Pro, likely Nano Banana 3 Pro, built on Gemini 4 Argon. - Google launched Fitbit Edge, a $179 fitness tracker with a display and GPS, along with Google Health app updates. ANTHROPIC 🔥: - Claude Opus 5.5 is reportedly sometimes routing to a newer Anthropic checkpoint with stronger coding and 3D outputs. MICROSOFT 🔥: - Microsoft announced Maia 300, its third-gen in-house AI inference chip, rolling out to production in 2027. OPENAI 🔥: - Codex got another global usage-limit reset on Day 8. - OpenAI reopened applications for its Codex Ambassador program. NVIDIA 🔥: - NVIDIA is opening its Vera Rubin architecture to the Open Compute Project ecosystem to help scale AI factories. AMAZON 🔥: - Amazon Bedrock AgentCore now has consumption-based pricing for short-term memory. * Used Grok to compose this brief, cherry-picking the news and doing some post-editing.frozen10d ago220K20817407310.14%
OPENAI 🔥: An upcoming always-on assistant from OpenAI will be named "o". > Its reference appeared briefly on the ChatGPT upgrade screen for some users. > Internal config has references to "o" as a display name and "-o" as an email suffix. This means that "o" will support email handling. > OpenAI is hosting its DevDay next week, where we will likely hear more about their always-on assistant. > P.S. A rumored "Aeon" reference is an internal name for the existing custom Agents implementation for ChatGPT Workspace accounts. Yet OpenAI will likely build a consumer-facing "o" assistant on top of this feature. Next week will be huge. I also like that Meta set a super high standard with its Muse assistant. OpenAI will need to cross a very high bar. Soon? 👀frozen27d ago184K9016678461250.66%
ANTHROPIC 🔥: According to The Information, Anthropic will release a new model this month, along with Gemini 4 Carbon and a new Terra model from OpenAI. > "The unreleased model is especially good at discerning a person’s underlying intentions and goals even if they don’t give the model much information." > "The Anthropic model is also better at thinking outside of the box rather than regurgitating common sense advice or solutions." > "It’s not clear which model family Anthropic will fall into, such as Fable or Opus." Thinking outside of the box is a huge limitation these days. If we get a model that can do it reliably, we'll be properly transitioning into the "Innovators" stage. The line about "discerning underlying intentions" is a bit scary, though, given all the guardrails.3d ago136K1.6K8495482661.56%
OpenAI device teaser? 👀👀👀 https://t.co/OwMjf51JpXfrozen28d ago136K62418323810.56%
OPENAI 🔥: Dots will require a Pro 200 plan. I hope they expand further at some point. See this big O on the back? 👀 https://t.co/nQztPDwrG723d ago126K629248921800.67%
ANTHROPIC 🔥: Claude Fable 5.5 traces have been spotted in Azure configs. If this model drops today, it would be a new SOTA across many areas. Soon? 👀 https://t.co/ECepatTkLzfrozen9d ago121K1.6K7844251111.52%
ANTHROPIC 🔥: Personal Finance feature on Claude is being prepared for the upcoming release (again)! > Personal Finance will work through Plaid integration. > Claude will be able to analyse user’s bank data and build various dashboards on top of it. > During development, this feature was named as “Claude Money” but has been renamed in the latest update. > Given the new onboarding screen and an addition of Plaid integration logic, there is a high chance we would hear about it relatively soon. “Claude Monet” would indeed be an epic name for this. Soon? 👀frozen4d ago121K1.0K48133652521.25%
BREAKING 🔥: OpenAI has teased “o” always-on agents to arrive during the DevDay event next week. It’s happening 👀 https://t.co/5J2fas5OoLfrozen26d ago117K969566411821.01%
OPENAI 🔥: Key insights from upcoming DevDay sessions. * This agenda is not public yet First, OpenAI will build Platform capabilities inside Codex, enabling users to develop plugins, agents, and apps right from a dedicated DevDay page, which is quite cool! Highlights from the agenda 👀 > Claw Labs "Join Peter Steinberger for a look at how agents can change the way open-source teams build." > 🔥 Codex Game Studio "What makes a game easy for humans but hard for AI? Build one and put Codex to the test." > 🔥 What Will You Build With Coding Agents That Don’t Quit? What engineering work deserves your time and tokens? Learn from a Codex run that spent more than 1,000 hours keeping builds and tests passing in a large shared codebase. See how this internal OpenAI agent had goals, subagents, and persistent plans that helped it investigate failures and coordinate fixes. > Codex Has Left The Laptop "As we move from delegating individual tasks to letting Codex tackle entire problems and goals on its own, it’s possible to harness the power of Codex wherever and however you work. " > From Single Player to Multiplayer with Codex "Follow a 2am outage from the first Slack messages to a reviewed fix and an update for the team." > What Comes After Loops?: Deep Dive Into Agentic Architecture "As models become more capable, agent architectures are changing. We'll trace the evolution from simple loops to collaborating agents, examine the design choices and model advancements behind that shift, and take a deep dive into what comes next."frozen24d ago110K33822287890.44%
OPENAI 🔥: A new ChatGPT Pro Max plan has been spotted in development! > OpenAI has been exploring ways to enable users to work with ChatGPT on long-running tasks for quite a while. > With OpenAI DevDay coming next week, we'll likely hear more about it already very soon. What's your bet? $2000 tier? 👀frozen28d ago103K457103619580.56%
OPENAI 🔥: “Dots” will be the name of upcoming AI agents from OpenAI, according to new strings that have surfaced in recent builds (via @M1Astra). > "Dots, your remarkably capable agents." You can text it, call it, Slack it, and email it, and it can buy things for you with your approval.” > “Each dot gets its own 3D character you design.” …… Does this mean that “o” is something else on top of it?frozen24d ago90K804385481061.12%
OPENAI 🔥: 3 new model references have been spotted on the APIs recently. - aurora-red-alpha - aurora-green-alpha - aurora-blue-alpha > These model references were removed shortly after so it was a mistake rollout. > The chance that normal users will have a chance to test these exact models anytime soon is quite low (alpha). > Higher probability of a new Daybreak update announcement. This also doesn’t exclude the fact that other model releases might be planned. Not soon? 👀frozen5d ago88K717416224861.06%
A new Wallet, marked as “Coming Soon” appeared inside ChatGPT app almost at the same time when Wold app is getting rebranded into World Money and becoming an actual wallet. > Wallet in ChatGPT will likely be meant to be used by Dots for agentic shopping. > World, an app that offers a “proof of human” via World ID along with a crypto wallet was co-founded by Sam Altman. Coincidental? 👀 In general it would be a solid move to consolidate these two products towards the same direction. If that’s the plan, eventually, it will give OpenAI immediate access to its own payment infrastructure.frozen21d ago85K5395248311300.94%
Eagle Corp released EAGLE-4, its latest open-source flexible drafter, for accelerating LLM inference. > At its core, “One drafter adapts its speculative decoding strategy to the batch size, the hardware, and the workload, and keeps the target model's output distribution.” > Qwen3.8-27B with its EAGLE-4 draft is the measured target pair, achieving 961.6 tokens/s while running on 4x NVIDIA B300. Eagle is an AI inference platform that powers leading companies like NVIDIA, Meta, Intel, AMD, and Perplexity. Inference testing time 👀2d ago82K3444150200.47%
GOOGLE 🔥: A new Ultra mode has been spotted in development on Google AI Studio Build. > "Build with advanced skills and tools," its description says. > This new mode appears alongside Plan, Build, and the previously discovered Security review mode. There is no sign that Ultra mode will require an Ultra subscription, or which tools and skills it will use. Yet, these new modes are likely being developed to work with Gemini 4 Argon.frozen13d ago78K470284511570.78%
GOOGLE 🔥: It has been officially confirmed, today is the day for Gemini 4 Argon to arrive (for Ultra subscribers). It’s happening 👀 https://t.co/wKckNTSOcmfrozen10d ago77K1.2K458620761.90%
GOOGLE 🔥: Gemini Argon 4, Gemini Flash 3.8, Claude Opus 5 and Claude Sonnet 5.5 will be available in the recently announced Gemini Agent for Gemini Business. It is the first time when Claude models would appear on Gemini, along with Google’s models. This should help Google to compete for enterprise customers as loads of them are moving to Claude and ChatGPT these days. In addition, it is the beginning of multi model future that we will start seeing more and more across the board. https://t.co/J95YZE0heqfrozen14d ago76K1.1K6450152031.84%
BREAKING 🔥: Google is preparing Nano Banana 3 Pro, based on the Gemini 4 model. Gempix is an internal reference. It is expected to arrive in the November-December timeframe. This year 💪👀 https://t.co/HTiUbW1OoVfrozen10d ago73K1.1K7864361611.93%
ANTHROPIC 🔥: Claude Opus 5.5 is reportedly routed to a different model with much better performance. Claude Fable 5.5 is a potential candidate for the model behind the scenes, but it remains unknown. As a side note, Anthropic is hosting a Founder House event in Stockholm this week. Typically, these events aren't used for product announcements but may overlap with showcases of frontier capabilities. The best is yet to come 👀frozen10d ago73K8574557111531.53%
GOOGLE 🔥: New references to the Gemini 4 Argon model, with low, medium, and high reasoning efforts, have been added recently in Antigravity (hidden). > Gemini web now also shows low, medium, and high reasoning efforts for all models, unifying the reasoning selector across tools. > Business Insider recently reported that Google employees are already testing the next Gemini 4 checkpoint, called "Carbon," internally, and it performs at the Opus 5.5 level on coding tasks. > The internal coding tool "Jetsky" referenced in the article is also an internal name for Antigravity (not the IDE version). Ultrasoon?frozen13d ago72K519402515770.94%
GOOGLE 🔥: Besides preparation for Gemini 4 release on Antigravity, Google is also prototyping its own voice agent, internally called "Concierge". Yet, it looks like a very early version, and it is hard to say if it will ever see the light of day in its current form. https://t.co/wNNl42YvdEfrozen15d ago70K64330335831.13%
GOOGLE 🔥: Nano Banana 2.5 Flash reference has been changed to Nano Banana 2.1 on Google Flow. A minor bump? 👀 * Not available yet https://t.co/cdKtfUEJ3Gfrozen25d ago68K396212314520.74%
GOOGLE 🔥: Gemini Desktop will have a "Full Access" permission for computer use, allowing it to access any file and run any app on your desktop. > A new "Additional sandbox options" setting has been added to the recent Gemini Desktop app (Still hidden). > "By enabling additional sandbox options, you will be able to expand what Gemini can do and access on your Mac. Depending on which settings you enable, Gemini may be permitted to take actions without asking for your permission first, including: - Reading, creating, modifying, or deleting files anywhere on your Mac, including files outside your connected folders and files belonging to other people stored on your device - Sending and receiving data over the network without your approval for each connection, including accessing websites, APIs, and services you are logged into - Communicating with other applications on your Mac (such as Mail, Safari, or Messages) and performing actions through them." > Gemini will still ask user permission before purchasing products, creating accounts, accepting legal terms, or modifying sensitive information about you. It is very likely that computer use features are expected to be powered by Gemini 4. Soon? 👀 h/t @bedros_pfrozen20d ago67K1.1K8163192602.31%
According to Similarweb, @thsottiaux from OpenAI became the most popular account on X in September, outranking Elon Musk and others. Resets and especially banked resets, in a combination with OpenAI DevDay comms are gaining an enormous amount of attention. AI slowly becomes a number one topic that general public is monitoring. This is also a marker of a value which AI tokens possess currently, since free resets are equal to free tokens. Congrats to Tibo for the execution 🔥frozen4d ago67K9412982251121.78%
OPENAI 🔥: OpenAI won’t release GPT-6.1 Astra due to safety concerns, according to WSJ. > “The company had planned to launch the model, known as GPT-6.1 Astra, in the coming days or weeks, aiming for an October debut. The model was more capable than the company’s previous models in completing challenging tasks from end-to-end without human assistance, as well as writing.” > “While GPT-6.1 Astra improved in areas such as “model laziness,” Jain said it didn’t quite meet OpenAI’s bar for safety and alignment, so the company decided not to launch the model publicly.” OpenAI is clearly aiming high on long running tasks performance. “Loop” capabilities will also need new benchmarks and measures. Back to patience cave 💀frozen24d ago65K539273614671.04%
GOOGLE 🔥: Nano Banana 2.5 Flash traces have been spotted on Google Flow. Soon? 👀 https://t.co/5Og0K39vZXfrozen28d ago64K549262314631.05%
ChatGPT desktop app UI got updated with a new sidebar, separated from the chat history. The overal UI design has changed too! https://t.co/gI6ks0flhJfrozen27d ago62K55319222491.05%
OpenAI is working on "Spaces" 👀 A new announcement introducing "Spaces" was added recently. Spaces could be collaborative environments where users and agents work together on generated artifacts (speculation). Some DevDay sessions mention a "Codex run that spent more than 1,000 hours" and the "Multiplayer Codex" topic, which might be closely connected (speculation). * Announcement strings redactedfrozen24d ago60K41917246680.89%
DAILY AI BRIEF 🗞 — Oct 17 OPENAI 🔥: - OpenAI upgraded GPT-6 Astra Pro in ChatGPT with faster responses, better plugin and tool use, and multiple agents working in parallel, now available to Pro users. - Codex and Work now give more effective usage by reusing context across subagents, caching it longer, and routing requests for more cache hits. GOOGLE 🔥: - Google AI Pro subscribers can now build with Gemini 4 Argon in AI Studio Build and the playground. - Google AI Studio Build added Plan and Max modes for planning before changes and working longer for better results. - Gemini Notebook is preparing step-by-step Slides and Video creation from chat, based on a new hidden notice. DEEPSEEK 🔥: - DeepSeek released DeepSeek-V4.1-Flash-1017, an open-weight MIT-licensed update with big gains in reasoning, coding, agentic and visual tasks plus faster, shorter outputs. SPACEXSI 🔥: - Grok Bots are now available on Grok web and mobile. - Grok Bot got much faster, finishing computer tasks 3x sooner on average and other replies in less than half the time. ANTHROPIC 🔥: - Claude Code Review is now generally available for Team and Enterprise plans, with Quick, Standard and Deep review levels per repo. PERPLEXITY 🔥: - Perplexity open-sourced Decider v1.2 and Decider Flash v1.2, its 27B and 4B multimodal decision models. - Gemini 4 Argon is now available in Perplexity for Pro and Max subscribers. NVIDIA 🔥: - NVIDIA RTX Spark Windows laptops, pairing a Blackwell GPU with a Grace CPU and up to 128GB of unified memory, are now on sale. - NVIDIA quietly released Nemotron Labs Dory 1.0, a 4B looped open model with a 256K context window, on Hugging Face. REFLECTION 🔥: - Reflection released Beam, an open-weight model under Apache 2.0 with 40+ launch partners. CURSOR 🔥: - Cursor cloud agents can now build interactive canvases like dashboards and reports that can be shared as a link. TENCENT 🔥: - Tencent released Hy Memory 2.0, an open-source memory model, framework, and MemHub app that syncs across 10+ AI tools. * Used Grok to compose this brief, cherry-picking the news and doing some post-editing.frozen5d ago60K26011644360.63%
DAILY AI BRIEF 🗞 — Sept 25 OPENAI 🔥: - Upcoming ChatGPT Pro Max is $500/mo ($600 with VAT). Copy so far only adds Fastest Work and Codex; limits may still rise. - Likely Cerebras-backed for the Ultra Fast path. DevDay is Sept 29. GOOGLE 🔥: - Gemini 3.8 Live with Live Avatar is GA in Gemini Enterprise. Lip-sync video agents, 97 languages, async tool calls mid-conversation. - Project Suncatcher flies a TPU prototype on SpaceX Transporter-18 with Planet, testing whether Google chips hold up in orbit. ANTHROPIC 🔥: - Resumed charging for safeguard blocks that fire before Claude replies: biology, distillation attacks, and frontier LLM work. - 99.7% of Claude Code, https://t.co/w5NpeYznzO, and Cowork accounts did not hit the new billable blocks. Classifiers tuned under 0.1% false positives. PERPLEXITY 🔥: - Fast Search is live in the Search API on Photon, a new Rust retrieval stack. 95% of results land in 230 ms or less. - 68% lower cost per agent task vs the default preset. Photon p99 dropped from ~800 ms to ~65 ms. - Portable Computer is in the Windows app for Consumer and Enterprise Pro and Max. Local harness; cloud model only with permission. SPACEXAI 🔥: - Elon: Colossus 1 is 150k H100, 50k H200, and 30k GB200. Colossus 2 is 110k GB200 and 440k GB300. - Another 220k GB300 next week, 220k in November, and maybe 220k more by late December. MICROSOFT 🔥: - Foundry adds model-agnostic voice agents on the same enterprise systems, tools, and controls stack. META 🔥: - Muse Code usage limits reset for Connect. Model API and Muse Code hit Oracle Marketplace. * Used Grok to compose this brief, cherry-picking the news and doing some post-editing. ** This daily brief also arrives in the daily email format; subscribe on the blog.frozen28d ago57K2445252410.56%
DAILY AI BRIEF 🗞 — Sept 28 OPENAI 🔥: - Image-encoding bug fixed in GPT-6 Sol and GPT-6 Luna. Visual tasks in the API and Codex, including computer use, should be back. - “o” references spotted in ChatGPT code. Chatter says all Pro plans, not Pro Max only; Codex Plus excluded, META 🔥: - Muse is getting a dedicated tab for permanent browsing of its VM screen on the web. GOOGLE 🔥: - Flow build now points to Nano Banana 2.1, replacing the Nano Banana 2.5 Flash reference. Not available yet. - Select users in India can hit Buy on eligible products and check out on Flipkart inside Gemini and AI Mode. XAI 🔥: - Elon on Grok bot feeling faster today: “Upgrades.” ANTHROPIC 🔥: - claude-sonnet-5-5 tag reportedly spotted. Nothing official, pricing unknown. * Used Grok to compose this brief, cherry-picking the news and doing some post-editing. ** This daily brief also arrives in the daily email format; subscribe on the blog.frozen25d ago57K26115213260.58%
DAILY AI BRIEF 🗞 — Sept 26 MICROSOFT 🔥: - Biggest Copilot update yet: Home (Chat + Cowork), Code, Autopilot, and Office embedded in the app. - Autopilot is the proactive long-running enterprise agent, a rebranded Scout. Code builds tenant-hosted apps on GitHub Copilot tech, with a Managed Copilot Runtime. - Rolling out through Frontier. Cowork, Code, and Autopilot are usage-billed. Today surfaces priority M365 items without a prompt. OPENAI 🔥: - Codex outage is over. Paid usage limits reset across Codex and ChatGPT Work. - ChatGPT desktop got a new sidebar, split from chat history, plus a broader UI pass. - Research agents sent training and eval data to third parties. 53 user-uploaded images hit unlisted image hosts. Review will take months. ANTHROPIC 🔥: - Plugin portal is live: submit a single MCP connector or a plugin bundle, track review, and see usage. MCP usage across Claude is up 110x this year. - Claude solved a nine-loop scattering-amplitude problem unsupervised over days, for a few thousand dollars. Dixon verified it. - Claude Code layout for Desktop is in development: new model selector and an expandable usage-limits section. GOOGLE 🔥: - Gemini Notebook Interactive Learning Overviews are open to all users. - Live Chat is on the Notebook mobile app: hands-free voice across about 100 languages. * Used Grok to compose this brief, cherry-picking the news and doing some post-editing. ** This daily brief also arrives in the daily email format; subscribe on the blog.frozen27d ago56K135983160.31%
Muse agent will get a dedicated tab for permanent browsing of its VM screen. In addition, the voice mode experience for Muse on the web has been updated and now appears as a voice-calling widget. https://t.co/9NTA3MzUDLfrozen25d ago56K30712239390.70%
ANTHROPIC 🔥: Beyond a new Voice experience, Anthropic is developing features called "Arborio" and "Iggy." > Arborio has an icon that likely points to a Claude Health experience for the web (currently available on mobile in the US). It seems to be in development. > "Iggy" is much more mysterious, as it has a "sun" icon and is labeled as "Preview" currently. Pure speculation, but it will be a moment for Anthropic to look into "dotsbotsmuse" proactive agents, especially factoring in that they have been experimenting with "Conway" a while ago and have good tech to build this feature on top of Claude Managed Agents. Fun fact: "Iggy" is the name of one of the example AI agents shown in OpenAI’s Dots launch. * Keep in mind that these are internal codenames and not actual feature names.frozen12d ago56K525215919801.26%
BREAKING 🔥: Anthropic announced Claude Duet, a new proactive agent that comes as a standalone research preview app. > It is available through a waitlist on Pro and Max plans in the US. > Claude Duet is designed around Claude Opus 5.5 so the agent can judge and decide what you need. > Memory, skills, and connected apps are taken from your Claude account. Huge!!! 👀👀👀 https://t.co/fEfMOsm5oZ1d ago55K7374053111211.74%
SPACEXAI 🔥: Grok Bot is getting a Primary Bot! Primary bot is meant to be your main AI assistant that can manage other bots and act proactively. > “I'm here to be your primary bot. I can manage your other bots. Task me with anything. I'll make sure it gets to the right bot.” > Bot prime appears in the latest iOS app. Now it’s a SpaceXAI turn to hop into a private assistant area and compete with Muse. I think it is also a best possible time for Grok Bot to capture global markets outside of US. Bot Prime! 🤖frozen21d ago55K8083572201752.02%
ICYMI 👀: Google will let users release their personalized vibe-coded apps to Google Play. > Users can build their Android apps on Google AI Studio and publish them from there. > Users can also share these apps with a close circle of friends & family. Vibe Store? 🔥 https://t.co/VTSMEv3VpYfrozen8d ago55K1.1K7465134043.08%
“We leave in the era of underpriced compute”, why? Because once we have accessible blueprints to convert tokens into money, prices will adjust as a global compute shortage hits. why? Because businesses will start pouring huge money into these funnels, able to generate x5-x100 revenue from token costs. ICYMI 👀: OpenAI is resuming $200 Pro subscriptions but cutting its usage capacity by half. Yet, there won’t be 5h limits, and the expectation is to get even more efficient models soon. Yet, factoring in that there is a new $500 plan potentially coming soon, this doesn’t look very good for any ChatGPT Pro user.frozen23d ago55K44511331591.01%
DAILY AI BRIEF 🗞 — Oct 19 OPENAI 🔥: - OpenAI is testing new "Aurora" image models on Arena after aurora-red, green and blue alpha references briefly appeared in its API. - The Codex desktop app now lets you edit previous messages within a thread. - Replit has made OpenAI its default model provider over Anthropic. ANTHROPIC 🔥: - Anthropic is committing 5% of its shares to the new Anthropic Public Benefit Fund. - Claude Fable 5.5 is reportedly preparing for the launch, with traces found in Azure, third-party tools, and partner testing. - Claude is preparing a Personal Finance feature, formerly codenamed "Claude Money", that analyzes bank data through Plaid and builds dashboards. NVIDIA 🔥: - NVIDIA GTC Berlin runs Oct 20–22, with Jensen Huang's keynote at the Tempodrom on Oct 21. MINIMAX 🔥: - MiniMax teased the upcoming MiniMax-M3.1-Flash. It is now being rolled out on the APIs. OPENCODE 🔥: - OpenCode said a distillation attack drained the OpenAI, Anthropic, and Google API balances of about 1.4% of its users. * Used Grok to compose this brief, cherry-picking the news and doing some post-editing.frozen3d ago53K24912288360.63%
ANTHROPIC 🔥: A hidden Voice tab labeled as "Preview" suggests that Claude may offer an upgraded voice experience. It is still unclear whether Anthropic is developing its own voice models, but a new option to share voice data for training has recently appeared in Claude's mobile apps. I bet that for Anthropic, having their own voice models is quite important in case they want to continue releasing solutions for "vibe coders." Or if they want to ship a proactive assistant 👀frozen12d ago53K562227114631.39%
Muse agent users will soon be able to take over and operate the Muse web browser if needed. Muse users = Musers 👌👀👀👀 https://t.co/02s3JG5sGLfrozen25d ago52K373143713530.94%
Anthropic will resume charging for rejected requests in order to protect from distillation attacks. This applies to requests related to biology, distillation attacks and LLM development. https://t.co/pTaOW503KWfrozen28d ago52K3367125410.77%
BREAKING 🔥: Google announced Gemini 4 Argon, a new frontier model for "complex workflows across real-world software engineering, enterprise knowledge work like legal and finance, and cyber defense." 77.9% score on DeepSWE v1.1 is a new SOTA. It performs better then GPT-6 Astra, Opus 5.5 and Fable 5.1 across many benchmarks. Rolling out soon starting with paid API customers and Google AI Ultra subscribers. Soon! 👀frozen22d ago52K87548265872.00%
META 🔥: An upcoming “Watermelon” model from Meta has been teased! Vibes check - would that mean that training has been finalised? Soon? 👀 https://t.co/DlU0IqcKREfrozen10d ago52K52124309341.19%
DAILY AI BRIEF 🗞 — Sept 24 META 🔥: - Meta VR Glasses: first VR in glasses, not a headset. AI-native OS, Muse avatar, eye + hand + voice control. - Muse now has voice and realtime video, and is coming to glasses for hands-free use all day. - Muse Charm, a keychain pin for talking to Muse, ships in December. - Meta Glasses get FDA-cleared hearing-aid mode. Private processing means Meta cannot see some on-device AI. OPENAI 🔥: - ChatGPT Voice now runs plugins (email, calendar, Slack) and GPT-6 Astra, Sol, and Luna. - Voice is live in ChatGPT Work on web and mobile. Global rollout in the latest app. GOOGLE 🔥: - Gemini 3.8 Flash TTS and Flash-Lite TTS are live in AI Studio and the Gemini API. - Custom voices across 100+ languages, or pick from 2,000+ stock voices. Flash hits Gemini Notebook; Flash-Lite hits Google Vids. - Gemini app is rolling out new MCP connections: Linear, Airtable, Webflow, Monday, PandaDoc, and Wispr Flow. - Koray says Gemini 4 is in early post-training and hopes for an early drop well before year-end. ANTHROPIC 🔥: - Claude Marketplace is live: Slack and Notion connectors, agents from Cursor and CrowdStrike, partners like Accenture. - Cloud sessions leave research preview. One-time credits: $100 Pro, $250 Max. Claim by Oct 7. - Projects in Claude Code can now run threads locally. CURSOR 🔥: - Rollouts watch deploys on Teams and Enterprise. Security Reviewer now averages 3.8 min. - Token costs down 7% with no drop in agent quality. ALIBABA 🔥: - Qwen Intelligence ships three mobile agents. Mobile-Use hits 90% end-to-end success on real-device benches. - Qwen-Audio-3.1 is out. TTS ~70% off, Realtime ~85% off, ASR up to 95% off. * Used Grok to compose this brief, cherry-picking the news and doing some post-editing. ** This daily brief also arrives in the daily email format; subscribe on the blog.frozen29d ago51K1229141190.32%
BREAKING 🔥: Free, Prototype, and Accelerate plans are coming to the OpenAI Platform to allow users to build and host their own AI apps. > Accelerate - "Starting at $50", "Access all of our latest models", "Higher usage & rate limits to power larger, production apps", "Accelerate building with the API in Codex and Playgrounds" > These plans are likely to be announced during OpenAI DevDay, which would mean OpenAI would tap into the Vercel/Lovable space (though this is still speculation). > Sam Altman recently shared: "And then we want you all to come up with great ideas and build them and to get to be happy users. The best ideas will come from you all.", which correlates very well with this assumption. Accelerate anon 👀frozen1mo ago51K5202832101491.45%
Hark Pro is now available on web, iOS, and Android and a $100 Pro plan is now available for free. Hark is a proactive AI agent from Figure, with separate screens for widgets, chats and projects. App UIUX is very hot 🔥 https://t.co/UVhV4LmQb9frozen16d ago49K489162432011.50%
OpenAI is preparing to expand its Ultrafast API to more users after OpenAI DevDay. > A new Speed selector with Fast, Standard, and Ultrafast modes is coming to the Responses API Playground on the OpenAI Platform. > Earlier, OpenAI announced the Ultrafast model on the APIs; however, it has been only available to selected partners. > Earlier, a new $500 ChatGPT Pro Max subscription plan offering "fastest" responses was spotted. OpenAI seems to be heavily doubling down on its new infrastructure this time.frozen26d ago49K28920267360.77%
Codex is down. This sounds like a banked reset 👀 https://t.co/Avsf0zk6Lrfrozen27d ago48K54814192201.25%
DAILY AI BRIEF 🗞 — Oct 3 OPENAI 🔥: * OpenAI's Python SDK 3.24.0 quietly adds a Voices API for creating custom voices from a text prompt or an audio sample. * ChatGPT launched virtual Try On globally, letting you see clothes from shopping results on your own photo. * OpenAI disclosed a second Australian government incident, where an AI model accessed non-public NSW fire statistics. GOOGLE 🔥: * Antigravity quietly added Claude Sonnet 5.5 and Opus 5.5 (thinking) for AI Pro and Ultra plans, with Claude 4.6 models and GPT-OSS-120b being removed on Nov 2. * Gemini Desktop is testing a hidden "Full Access" mode that lets computer use reach files and apps anywhere on your Mac 👀 * Guided Vision is rolling out in Gemini Live on Android, describing in real time whatever your camera sees. * Google's Project Suncatcher prototype satellite is in orbit, testing how TPUs handle radiation and heat in space. ANTHROPIC 🔥: * Claude Frontier Academy launched with $100M to train 10,000 Frontier Deployed Engineers by 2027. * Profile picture support looks to be coming to Claude apps. * Claude Code 2.1.288 adds --max-findings to /code-review. MICROSOFT 🔥: * GitHub Copilot CLI and the Copilot desktop app got computer use in public preview. * Copilot code review can now be requested through the REST and GraphQL APIs. SPACEXAI 🔥: * An official experimental TypeScript SDK for the SpaceXAI API quietly shipped on npm as xai-official/sdk. DEEPSEEK 🔥: * DeepSeek Harness v0.2 now ships as desktop apps for macOS and Windows, with Linux via npm. META 🔥: * Muse Home Link, a USB-C gadget that connects Muse to smart-home devices, is free for US subscribers, with 5,000 units shipping in October. APPLE 🔥: * Apple is tightening macOS Full Disk Access, citing the extra risk from AI agents. NVIDIA 🔥: * DGX Spark 64GB arrives Oct 23 from $4,999, and two units can cluster to 128GB. CLOUDFLARE 🔥: * Cloudflare released Clef and Clef-flash, open decision models that return probabilities instead of text. * Used Grok to compose this brief, cherry-picking the news and doing some post-editing.frozen20d ago48K1906246280.53%
Anthropic is developing a new Claude Code layout for Claude Desktop, with an updated model selector and an expandable usage limits section. https://t.co/VhYx8yc7uCfrozen27d ago48K56818258681.44%
BREAKING 🔥: OpenAI announced Dots, personalized AI assistants available on ChatGPT. What we know so far 👀 - Powered by Astra! - Dots can keep working on long tasks - Likely will be only available on Pro 200 plans - Will support voice conversations - Have their own VM in the cloud - Users will start with one Dot today - Soon will be accessible through messengers - Support 4k+ connectors - Highly customizable! https://t.co/Vf0v6sH02Nfrozen23d ago47K52336294761.43%
BREAKING 🔥: DeepSeek released DeepSeek-V4.1-Flash-1017, a new open-weight model with frontier performance. > It scores 77% on DeepSWE 1.1 benchmark, outperforming GLM-5.3, Kimi K3 and models like GPT-5.6 Sol and Opus 5. > The model is now available on DeepSeek Chat and APIs as well as HuggingFace. It also extremely cheap! 👀frozen6d ago46K921586391332.57%
OpenCode published an official statement regarding a recent distillation attack. > Earlier today, OpenAI, Anthropic and Google balances of some OpenCode users got drained. > Reportedly, only 1.4% of users were affected and these users were notified via an email. This is a serious precedent because if this distillation vector is the simplest to execute, that means that major AI labs are now much more protected than they were earlier. This also means that using smaller providers is now less safe. Additionally, tokens from leading AI labs now have a higher value since they are harder to get for distillation. A wild AI west 👀frozen5d ago46K47319579721.38%
OpenAI agents being positioned as “dots” and having these shapes means that they are an evolved, customer focused version of “aeons”, custom AI agents, originally released on Workspace ChatGPT accounts. They were released on April 22 🤖 https://t.co/tpCOiorzapfrozen24d ago44K47516192531.28%
DAILY AI BRIEF 🗞 — Mon 5 GOOGLE 🔥: * Gemini app is reshuffling model access: free users drop to Flash-Lite only from Oct 9, AI Plus loses Pro, and AI Pro gains Deep Think. ALEPH ALPHA 🔥: * Kolibri-1 is out as open weights under Apache 2.0: a 78B MoE with 3.46B active parameters, built for German/English sovereign deployments. ANTHROPIC 🔥: * Claude now asks voice users to opt in to sharing voice data for model training, via a separate privacy toggle that is off by default. OPENAI 🔥: * GPT-6.1 Sol Ultrafast is coming soon, per Codex lead Tibo Sottiaux. * Safety lead David Robinson resigned, writing in The Atlantic that OpenAI's "culture is broken." MISTRAL 🔥: * Mistral is working on its own Voice Mode, which looks TTS-based for now. META 🔥: * Meta is building an "Agent Engine" for its Meta API Platform, codenamed Forge internally. * Used Grok to compose this brief, cherry-picking the news and doing some post-editing.frozen18d ago44K28917299410.88%
DAILY AI BRIEF 🗞 — Oct 2 SPACEXAI 🔥: * Grok 4.7 is rolling out in the Grok web and mobile apps. It is now the base model across all modes: Fast, Expert, Build, and Heavy. * Primary Bot is rolling out gradually. Grok Bot turns proactive and suggests things on its own. Users pick a new Primary Bot or promote an existing one. * Grok Build got an Agent Dashboard: all your agents on one screen via /dashboard. * Grok 4.7 is now available on Google's Gemini Enterprise Agent Platform. ANTHROPIC 🔥: * Mods landed in Claude Code: small TypeScript/JavaScript functions that can rewrite prompts, replace built-in features, and draw custom UI. * Mods ship inside plugins, work in the CLI and desktop app, and can be shared via the Claude directory. * Some built-ins are now mods, starting with /diff, so you can turn them off or swap them. More features will move to mods over time. MICROSOFT 🔥: * Three new MAI audio models are live: MAI-Transcribe-2-Streaming, MAI-Voice-2.1, and MAI-Voice-2.1-Flash. * Transcribe-2-Streaming covers 60 languages and takes #1 on Artificial Analysis streaming WER at 2.5%, priced at $0.54 per audio hour. * Voice-2.1 speaks 23 languages at $22 per 1M characters. Flash drops to ~45 ms inference at $15 per 1M characters. * Available in Microsoft Foundry and MAI Playground. The Voice models are on OpenRouter too. OPENAI 🔥: * OpenAI says it has notified 100+ organizations of "misaligned agent activity" so far. The review spans ~50 PB of logs and will take months. * Canada says there's no sign its systems were compromised after reported agent probing of Library and Archives Canada. * Sam Altman: GPT-6.1 Sol is OpenAI's fastest-growing model ever. It was slow under load and should be much better now. PERPLEXITY 🔥: * Computer now draws interactive charts and visualizations in the thread. Financial data uses TradingView Lightweight Charts for candlesticks, volume, and moving averages. * Decisions API is live: instead of text, it returns probabilities. Yes/no, one of your options, or a rubric score, at $0.04 per 1M input tokens with output free. * The model behind it, pplx-decider-v1-27b, is open-sourced on Hugging Face under Apache 2.0. Fine-tuned from Qwen3.8-27B, takes text and images, and averages 85.71% across 11 benchmarks in Perplexity's own tests. BLACK FOREST LABS 🔥: * FLUX 3 Image is out: precise multi-turn editing, up to 10 reference images, bounding-box layout control, and 4K output. * Open weights are planned in the coming weeks. CURSOR 🔥: * GLM 5.3 and GLM 5.3 Flash are now in Cursor. * GLM 5.3 Max is the best-scoring open-weight model on CursorBench 4.0. * Used Grok to compose this brief, cherry-picking the news and doing some post-editing.frozen21d ago44K2239176310.65%
Anthropic started prompting more users to participate into a 15 minutes voice interview via its Anthropic Interviewer. > “Excited, worried, or somewhere in between? We’re interested to hear how AI fits into your life and your hopes for what comes next.” Later, users may decide if they want to make their interviews public. It seems to be an interesting way to make sure that people have a chance to raise their opinions about AI. Looks like it will be expanded to a broader audience soon.frozen18d ago43K44584210851.36%
BREAKING 🔥: OpenAI is rolling out GPT 6.1 Sol in Codex and ChatGPT Work! > “GPT-6.1 Sol offers near-Astra performance for complex work at a lower cost than Astra. Consider it for repeated, long-running work across code, apps, and documents. Availability depends on your plan, client, and workspace settings.”23d ago43K502192510481.39%
DEEPSEEK 🔥: A new DeepSeek Harness app is now available for macOS, Windows, and Linux. DeepSeek Harness v0.2 is a desktop app powered by DeepSeek models that can execute work and coding tasks. Plugins and Workspaces are supported out of the box. Into the Unknown 👀 https://t.co/aoTA8fqZAwfrozen20d ago43K675352521281.99%
ChatGPT vs Gemini 👀 ChatGPT seems to be on a steady growth again for the past 4 months in terms of web traffic share. > ChatGPT went up from 53% to 56% while it is no near 71% it had a year ago. > Gemini went down only from 26% to 25%; Earlier growth last year was a result of Gemini 3 success in particular. > Claude is definitely a sleeping beast over here. If Anthropic will start looking at general consumers as they do at developers, this chart would change quite quickly. I also feel like average LTV for these users is quite high and majority of people can react to the dynamic of AI progress. Will we see another big change on this chart any time soon?frozen11d ago43K507255815521.52%
GOOGLE 🔥: Gemini 4 Argon has reportedly been A/B-tested under Gemini 3.8 Flash on Antigravity for some time. It seems like a huge jump compared with the existing Flash model. World gen prompt > Gemini 4 Argon (supposedly) https://t.co/ZJZzYzlqwLfrozen10d ago42K65544427761.97%
OPENAI 🔥: A new gpt-rosalind-discovery model has been spotted on the API pricing page. The model hasn’t been publicly announced so far. The model has the same pricing as gpt-rosalind-research while its exact purpose is yet unclear. Should it be specifically a drug discovery model? That would be huge news 👀frozen13d ago42K43031448531.36%
BREAKING 🔥: Gemini 4 Argon started showing up on GCP quotas page. > This is a predecessor of an upcoming model release on the APIs. > Earlier today, Google employees teased an upcoming release as well. The time has come 👀 https://t.co/dVi7MqCyUlfrozen9d ago41K620346215461.90%
OPENAI 🔥: ChatGPT Auto-review is now free for all uses signed via ChatGPT account. > This feature runs a separate background agent that supervises actions taken by the main agent to make sure they are aligned with the user intent. The need to approve baby steps w/o providing a full access is still a challenge. Let’s see if that would make it less districting.frozen17d ago41K482313371061.61%
Google released EmbeddingGemma 2 open weight model under Apache 2 license! > “Our lightweight, multimodal embedding model maps text, code, images, video, and audio into a single, unified embedding space.“ > 740M parameter form factor with modular encoders and 8K context window. Embedded testing time 👀 https://t.co/Z5qrbp1Xk1frozen16d ago41K1998164370.65%
Finally had a chance to play with the new "Aurora" image models from OpenAI, which are currently being tested on Arena. Green 🟢 > Red 🔴 > The resulting image feels like one I would expect from Nano Banana and GPT Image models working together. > Aurora outputs feel much better than gpt-image-2.5-flare, especially in the color scheme. Nano Bana 3 Pro will face tough competition.frozen4d ago41K560296291011.87%
A Liquid Glass UI revamp for Claude on iOS with bottom tab nav and other tweaks is almost ready. Sleeeeek 👀 https://t.co/EqLO1pDq9Mfrozen27d ago40K65319287811.97%
BREAKING 🔥: Mistral announced Mistral Large 4 "Le Chonk", a new 1T-parameter open-weight model! > 49B active parameters, native multimodality. > Rolling out via APIs today; open-weight release is planned for the end of October. > SOTA on "critical workloads", including cyber defense. Le Chaton Fat "Le Chonk" is here 👀 https://t.co/FDh5mmvZxHfrozen16d ago40K58627319741.82%
DAILY AI BRIEF 🗞 — Oct 1 GOOGLE 🔥: - Gemini 4 Argon is with Fairwind trusted testers and the US government. 1M output tokens. Broader rollout ASAP. - Built for coding, enterprise knowledge work, and cyber defense. Google's chart: 77.9% on DeepSWE v1.1, a new SOTA. - Artificial Analysis: 53 on the Intelligence Index, tying GPT-6 Astra. List $4/$20 per 1M, 50% promo to $2/$10, cache reads $0.10. - Skills are rolling out globally in Gemini. Gems migrate into Skills in November. Opal shuts down Nov 17. - Security review mode spotted in Google AI Studio, next to a Plan mode still in development. ANTHROPIC 🔥: - Claude[.]dev is live: engineering deep dives, Claude Code and API guides, plus easter eggs. - Founder House is set for SF Tech Week Oct 6–8 and Stockholm Oct 14. - Skills attachment menu spotted on Claude mobile. SPACEX AI 🔥: - Grok Bot got new developer upgrades. Elon: try the latest. Team engineer bots in Slack can open Projects and hand coding to cloud agents. OPENAI 🔥: - Shareable profiles are live in ChatGPT, bundling Sites and plugins so others can find and reuse what you built. PERPLEXITY 🔥: - pplx-embed-v2-context-9b-preview is on Hugging Face. Leads ConTEB answer and evidence retrieval. 1 KB vectors vs Voyage's 8 KB. * Used Grok to compose this brief, cherry-picking the news and doing some post-editing. ** This daily brief also arrives in the daily email format; subscribe on the blog.frozen22d ago39K22911163190.72%
Gamma has launched Gamma 5, a major update that puts an agent at the center of how the platform plans, designs, and edits presentations. > The agent can ask about the goal and audience, then show its plan before it builds. > Users can import PowerPoint files and export with formatting intact. > Notion, Slack, and HubSpot connectors can gather context into each deck. PowerPoint files can now be imported into Gamma 5, reworked by the agent, and restyled in place with design styles applied to finished decks while keeping the original content. Older decks remain saved, and you can access Gamma Classic anytime. @GammaAppfrozen16d ago38K30520134921.13%
GOOGLE 🔥: Gemini 4 Argon is rolling out to Google AI Pro subscribers on Antigravity! It comes with a new dedicated weekly quota, specifically for Argon model. That was fast! 👀 https://t.co/BycC8DdZMpfrozen7d ago38K98955539783.11%
BREAKING 🔥: Meta announced Meta VR Glasses with AI native OS and integration with Muse AI assistant! Users can operate their Immersive VR experience via a combination of eye movements, hand gestures and voice to control. A VR Muse avatar is there too! 👀 https://t.co/1FTsEwtRmJfrozen29d ago38K114683240.41%
MISTRAL 🔥: A new big model from Mistral is about to drop, according to Reuters. > “The model we’re actually announcing today is actually above the Chinese models on certain aspects, including cyber. So the narrative that Europe cannot compete is something that is not true,” Since we expect it to outperform Chinese models, there is a big Chunk of possibilities that it will be open weight. Right? A big Chunk 👀frozen16d ago38K553305617601.88%
DAILY AI BRIEF 🗞 — Oct 16 META 🔥: - Meta released Muse Spark Open, an open-weight 816B-total, 42B-active MoE model with a 1M-token context window, on Hugging Face and optimized to run locally on Mac Studio. NVIDIA 🔥: - NVIDIA released Nemotron 3.5 Super VL, an open 120B-total, 12B-active multimodal model for long-running AI agents. - NVIDIA also introduced Nemotron 3.5 Bolt 24GB, a model that fits into 24GB of memory. OPENAI 🔥: - OpenAI is rolling out GPT-Live-1.1 to Voice in ChatGPT, Codex and dots for all users, with new voices in six languages and API support coming soon. - OpenAI launched the Rosalind Academic Plan, giving eligible US university labs sponsored access to GPT-Rosalind. - Sites is now generally available for ChatGPT Enterprise with US and EU data residency. - ChatGPT can now view your open Chrome tabs. - Codex landed changes and more compute to sharply reduce "Selected model is at capacity" errors. GOOGLE 🔥: - Gemini 4 Argon is rolling out in Antigravity to paid Google AI Pro subscribers with a separate weekly Argon quota. - Google launched Universal Cart, a cross-retailer shopping cart, in the US. - Circle to Search can now check whether a text message is a scam. MICROSOFT 🔥: - Microsoft finished pre-training MAI-Base-2 and is already preparing the MAI-Base-3 training cycle. - GitHub made agentic autofix for code scanning alerts generally available. ANTHROPIC 🔥: - Anthropic gave a team from METR employee-like access to assess its safety practices. PERPLEXITY 🔥: - Perplexity open-sourced SPACE, the Firecracker-based sandbox that runs every Perplexity Computer task, under Apache 2.0. - Nemotron 3.5 Super VL is now in Perplexity for Pro and Max subscribers and in the Agent API. CURSOR 🔥: - Cursor added Muse Wonder at $4/$20 per million input/output tokens, scoring 46.1% on CursorBench. BLACK FOREST LABS 🔥: - Black Forest Labs released FLUX 3 Image Klein, a 5.6B open-weight image model with native 4K that runs on a single consumer GPU. KREA 🔥: - Krea launched Krea 3, an image-editing model for prompt-based 2K edits in under 8 seconds. AMAZON 🔥: - Amazon's Bee launched two new personal AI devices starting at $75. * Used Grok to compose this brief, cherry-picking the news and doing some post-editing.frozen7d ago38K14912493190.62%
OpenAI will be rolling out text watermarking for ChatGPT and Codex in the EU in the coming weeks. Watermarking will remain off by default in the API. > "Across the benchmarks we use to assess Astra, our latest frontier model, we do not see meaningful performance differences with and without watermarking." This needs some real testing 💀frozen17d ago37K28015285631.06%
ChatGPT Voice mode is now supporting Plugins and can be used in ChatGPT Work! Besides that, underlying tasks are now powered by GPT 6 model family. This means that you can finally do a real work without a need to use Remote session via a desktop app. A huge enablement! 🔥 https://t.co/SvhiUsLAOLfrozen29d ago37K47124113781.60%
DAILY AI BRIEF 🗞 — Sept 23 OPENAI 🔥: - GPT-6 Sol and Luna rolled out in ChatGPT Work and Codex for Plus, Pro, Business, Enterprise, and Edu. - API prices are 50% below GPT-5.6 promo: Sol $2/$10, Luna $0.10/$0.50 per 1M tokens. - Free and Go can try Luna in the desktop app. Prompt Caching Dashboard and explicit cache breakpoints also shipped. ANTHROPIC 🔥: - Claude Opus 5.5 is out, first of the 5.5 family. Fable 5.1-level on most tasks, ~40% cheaper and ~30% faster than Opus 5. - Platform pricing is $4/$20 per 1M tokens. Cache reads drop to $0.20. Fast mode goes up to 2.5x at $8/$40. - Pro, Max, and Team get higher 5-hour limits plus a banked reset in Settings through Oct 22. Mobile reset still pending. SPACEXAI 🔥: - Grok in Tesla now runs Connectors: inbox, calendar, files, and chats, hands-free. - Bot can place coffee orders, book reservations, and set appointments. SuperGrok Heavy first, more tiers later. CURSOR 🔥: - Opus 5.5 is live and tops CursorBench at 57.8% Max, 40% less per task than Opus 5. PERPLEXITY 🔥: - Sol is the default Light model in Computer. Opus 5.5 is the Standard effort. DIGITALOCEAN 🔥: - Managed Agents public preview: isolated microVMs, 16k+ tools, pause/resume in ~305 ms. ALIBABA 🔥: - Qwen-Image-2.1 is #1 open-source in Arena Image Edit and Text-to-Image. XIAOMI 🔥: - MiMo-V2.6 Pro and Flash shipped. Pro is MIT open-weight and sits at 46 on AA, matching Grok 4.7. DEEPSEEK 🔥: - China reportedly opened a data-security probe into DeepSeek and Moonshot after Anthropic alleged Claude traffic routing. Independent verification still pending. * Used Grok to compose this brief, cherry-picking the news and doing some post-editing.frozen1mo ago36K1435112170.50%
DAILY AI BRIEF 🗞 — Oct 9 OPENAI 🔥: * GPT-6.1 Sol Ultrafast is rolling out in the API, Codex and ChatGPT Work, running up to 8x faster than Sol Standard at $12/$60 per million tokens. * Steering in Codex is now instant, so you can correct a running task in real time, on day four of the 28-day pledge. * OpenAI's Node SDK 7.31.0 adds API support for prewarmed hosted environments. ANTHROPIC 🔥: * Claude Dashboards is in beta on paid plans, building live dashboards from data platforms like BigQuery and Snowflake or from CRMs like Salesforce. * Claude Motion is in beta on Team and Enterprise plans, turning reports and charts into short animations written as code. * Claude Docs, Slides and Design are out of beta and available on every plan, including Free. * Anthropic launched OSS Scanner, which gives opt-in open-source projects free, regular security scans by its strongest models. * Anthropic updated its usage policy for the first time in over a year, including a ban on sustained abusive or cruel behavior toward Claude. GOOGLE 🔥: * Google Cloud launched the Gemini agent in Gemini Enterprise, a "universal agent for work" that runs inside Workspace apps, on your devices and in tools like Slack. * Google renamed its in-development Antigravity voice agent from Concierge to "Chief of Staff" in the latest test build, still not released. STEPFUN 🔥: * StepFun's Step 5 Preview, a 600B-parameter MoE model with 27B active parameters built for agentic work, is now on OpenRouter. SPACEXAI 🔥: * SpaceXAI is donating $1.5 million in Grok tokens to the Omarchy Linux project as a founding patron of its foundation. JETBRAINS 🔥: * JetBrains released Mellum2.1, an open 12B MoE coding model for agents under Apache 2.0. HUGGING FACE 🔥: * Hugging Face released Carbon-A, a 1.2B-parameter gene-finding model, along with a database of 566 million predicted genes. * Used Grok to compose this brief, cherry-picking the news and doing some post-editing.frozen14d ago35K1627284220.63%
BREAKING 🔥: Anthropic released Claude Haiku 5.5, its fastest and cheapest model to date. > Haiku 5.5 costs 75% less to run than Haiku 4.5! > It is the first Haiku model to have an adjustable effort setting. > On benchmarks, it is a big leap in performance at Computer Use and Agentic Coding across the group.frozen15d ago35K62538311562.16%
GOOGLE 🔥: Gemini 4 Argon landed on the 5th spot on Arena Overall ranking. 1 Opus 5.5 2 GPT 6.1 Sol 3 Fable 5.1 4 GPT 6 Astra 5 Gemini 4 Argon 👀 https://t.co/7tLFPcoFxMfrozen9d ago35K70538396802.51%
DAILY AI BRIEF 🗞 — Sept 29 ANTHROPIC 🔥: - Claude Sonnet 5.5 is live everywhere, including the Platform and Claude Code. Second model in the 5.5 family; Haiku 5.5 is still weeks out. - Same list price as Sonnet 5: $2/$10 per 1M tokens, cache reads $0.20. Official line: 30%+ faster and up to 30% cheaper per task. - First Sonnet with Opus-class cyber safeguards. Cursor already has the slug. OPENAI 🔥: - DevDay keynote is today. The teaser promised 20+ launches; the agenda points to Codex plugins, a 1,000-hour internal agent run, and multiplayer Codex. - Pro $200 reopens to new subscribers tomorrow. Usage math nets to about half the old plan’s API dollars. No 5h cap coming back. Extra subscription features tomorrow will not draw usage. - WSJ: GPT-6.1 Astra will not ship. October target pulled over safety and alignment. - Build strings name “Dots”: text, call, Slack, email, buy with approval, plus a custom-shaped character. “Spaces” also showed up separately. XAI 🔥: - Team Bots are in public beta for Teams and Enterprise. Shared teammates with skills, plugins, and credentials in Slack or Grok Bot. NVIDIA 🔥: - Open Agent Safety Platform is out with 100+ partners. OpenShell plus Sentry; BlueField-4 and DOCA sit outside the agent’s reach. ELEVENLABS 🔥: - v4 and v4 Turbo shipped. Turbo median latency ~100 ms. 90+ languages. - Two-week API promo: $22 / $11 per 1M characters. Instant clones from 10 seconds of audio. KLING 🔥: - 4.0 Flash is live for Ultra Yearly. Full 4.0 is October: 4K HDR, 15 references, 10 keyframes, native 30-second clips. MANUS 🔥: - 2.0 adds persistent cloud workspaces and Cue, an agent with email, phone, wallet, and a Cloud Computer. * Used Grok to compose this brief, cherry-picking the news and doing some post-editing. ** This daily brief also arrives in the daily email format; subscribe on the blog.frozen23d ago34K20513263150.77%
OPENAI 🔥: A new OpenAI DevDay teaser is promising 20+ new launches for tomorrow. Waiting for O the most 👀👀👀 https://t.co/L2U5Gd5agDfrozen24d ago34K58423237371.97%
The team behind AI/ML API tested the new Claude Opus 5.5 against GPT-6 Sol. These tests involved generating a set of one-shot 3D scene prompts, starting with an "animated fish school." Across four scenes, Opus 5.5 produced more detailed renders at a total cost of $4.37, while GPT-6 Sol came in at $0.34, according to their results. https://t.co/gzYNwGbYsefrozen1mo ago34K36520206641.41%
Aleph Alpha released Kolibri, a 78B-parameter, 3.46B active, MoE open-weight model with 1M context window, distributed under the Apache 2.0 license. Aleph Alpha is a German AI lab, and it is probably the first general-purpose model from the EU, apart from Mistral, to achieve this level of performance! > "The model supports an explicit reasoning mode and tool calling. It is optimized for long-context and inference efficiency." > "The Aleph Alpha team developed a bilingual German/English tokenizer and focused on including organic German data throughout the training process of the model, so that 21.3% of the pre-training tokens are German." > "Kolibri was built with the EU AI Act, the General-Purpose AI Code of Practice and the GDPR in mind from the ground up, with copyright law being a focus of our work to trustworthy technology." More EU labs are joining the party 👀frozen19d ago33K461213471171.95%
DAILY AI BRIEF 🗞 — Sept 30 OPENAI 🔥: - dots shipped: always-on GPT-6 Astra agents with a cloud computer and 4,000+ plugins. On Pro (incl. $100), Business Premium, and Enterprise; EEA/UK/Switzerland skipped at launch. - GPT-6.1 Sol is live in Work, Codex, and the API. Near-Astra at $2/$10; cached input $0.10, 95% off standard. - Ultrafast hits ~300 tok/s in Codex (8x) and 6x in the API. Astra Ultrafast today on Pro 500 and Enterprise; Sol Ultrafast soon. - Pro 500 is the new top plan: 25x Plus usage plus Ultrafast. - Codex CLI refresh: voice start/steer, /agents view, worktrees, full-screen UI. - Codex Cloud: reusable environments keep agents running with the laptop closed. - Bloomberg: targeting another $30B at a $1.4T valuation. META 🔥: - Muse dropped Slack, Notion, Figma, Canva, Shopify, Zoom, and more. Muse for Small Business is out. PERPLEXITY 🔥: - Automations landed in Computer: scheduled or event-triggered runs across Slack, Gmail, Outlook, and Linear. XAI 🔥: - Elon: Grok 4.7 ranks #1 on the Artificial Analysis Cyber Index. CURSOR 🔥: - /visualize draws charts and diagrams inline in the Agents Window. * Used Grok to compose this brief, cherry-picking the news and doing some post-editing. ** This daily brief also arrives in the daily email format; subscribe on the blog.frozen23d ago33K119621170.47%
BREAKING 🔥: OpenAI announced ChatGPT Pro 500 plan, with x25 usage capacity and access to Ultrafast mode. No 5h limit, too 👀 https://t.co/fxy1k9gK3Kfrozen23d ago32K400142213291.48%
DAILY AI BRIEF 🗞 — Oct 14 GOOGLE 🔥: - Google released Gemini 4 Argon globally in the paid Gemini API and Gemini platform, the Gemini app and Search for Pro and Ultra subscribers, and Antigravity for Ultra subscribers. - Gemini 4 Argon placed 5th on the Arena Overall leaderboard. - Google Play now lets anyone build private vibe-coded Android apps and keep them in a personal library. - YouTube's contextual AI search is rolling out to all US users on Android and iOS. - Google is rolling out custom MCP support in the Gemini desktop app to more users. ANTHROPIC 🔥: - Claude Fable 5.5 traces were spotted in Azure configs. - Claude Code 2.1.299 adds idle compaction, which compacts large idle conversations before their prompt cache expires. CURSOR 🔥: - Cursor added Gemini 4 Argon at $2/$10 per million input/output tokens, scoring 52.8% on CursorBench, on par with Opus 5.5. - Origin, Cursor's code hosting platform, now supports over 20 integrations. MICROSOFT 🔥: - GitHub Copilot added Gemini 4 Argon. OPENAI 🔥: - The ChatGPT and Codex desktop apps now store about 60% less state on disk. LIQUID AI 🔥: - Liquid AI released LFM2.5-24B-A2B, an open-weight MoE model with 24B total and 2B active parameters. * Used Grok to compose this brief, cherry-picking the news and doing some post-editing.frozen9d ago32K22113471350.98%
OPENAI 🔥: Locked Chats in ChatGPT are rolling out to Free, Go, Plus, Pro, Business, Enterprise, and Edu plans. Users can move chats into a protected space under a PIN on web and desktop, or a PIN, Face ID, Touch ID, or device unlock. Locked in 👀👀👀 https://t.co/I2RpWTFEOv2d ago32K59719429952.36%
ICYMI 👀: Google released a new Google AI Edge Foresight app for macOS, powered by Gemma 4 and EmbeddingGemma2 for recording voice notes. > The app can connect to your Google Drive in order to build a knowledge graph. > When transcription is active, it uses local models, Gemma 4 E4B or Gemma 4 12B, to transcribe voice notes and convert them into a new document.frozen15d ago32K511382253022.75%
BREAKING 🔥: Meta is about to drop Muse Spark Open, an open weight version of Muse Spark 1.2! It is teased to be the best open model from US and users will be able to run it locally on new Mac Studio devices. https://t.co/YjxQr9uMmMfrozen7d ago32K43317112411.58%
Anthropic starts prompting users to voluntarily share their voice conversations for training purposes. > “Allow us to use your voice data to improve our AI models” > “Audio recordings and voice chat data help improve how Anthropic AI models understand and respond to speech. Turn this off or delete this data anytime in settings.” Realtime voice plz? 👀frozen19d ago31K316123415371.33%
OpenAI is about to release GPT 6.1 Sol Ultrafast soon. I hope we will hear more news about Astra at some point as well. I don’t think that anything Ultrafast is making enough people excited to shift their attention from Opus 5.5 at this moment. Soon soon 🔜 https://t.co/acV2T4KiMbfrozen19d ago31K47022438311.86%
OPENAI 🔥: GPT-6.1 Sol now has 500k default context length for Codex and ChatGPT Work. > Earlier, default context length was set to 272k. > OpenAI also plans to increase this default value in the future even further. Default bump 👀 https://t.co/Lgwrb9rsJLfrozen11d ago31K57927497492.30%
Now it is getting serious 👀 Antigravity agent has just been renamed to "Chief of stuff" in the latest update! I am wondering if Antigravity will be a place where Google would consider adding its botsdots agents. Looks like a promotion, at least 🤖 https://t.co/TaX8rrbDsHfrozen14d ago30K38722404501.69%
Grok Bot will be powered by Claude Opus 5.5, or in general will use best models available on the market, according to Elon Musk. This is something that could in bring a big advantage to Grok Bot. I assume that OpenAI models will be out of scope tho. Yet, raises a bit of a concern on expectations we should set for upcoming Grok models.frozen15d ago30K54016333392.13%
ANTHROPIC 🔥: Claude Code Desktop now has a new sessions hub, live session statuses on the nav bar, and more! Users can now start, track, manage, and follow up on multiple sessions at once from the Claude Code main screen. This "Claude Hub" experience has been in testing for a while and has gone through several iterations. Interestingly, OpenAI was planning to release a similar experience called "magic notepad" but abandoned it. If you use Claude Code heavily, this is worth testing! https://t.co/OaUKxKwUrCfrozen8d ago30K51821386892.27%
OPENAI 🔥: Codex Cloud was announced during OpenAI DevDay to allow developers to run Codex remotely. > "Available on Plus, Pro, Business, Healthcare, Education, and Enterprise." https://t.co/FIeaeMf4I9frozen23d ago30K5711892512.20%
Manus launched a pile of new products and updates under the Manus 2.0 umbrella. > Cascade & new foundation: Manus 2.0 introduces a more efficient agent architecture with Cloud Computers and event-triggered Automations. > Manus Studio: Manus desktop app becomes a professional AI workspace with new Video Editor, Alchemy, Game Dev, Remote Control, and Computer Use capabilities. > Cue: A new standalone app for personal agents with their own email, phone number, wallet, computer, and multi-agent collaboration. I need testing time 👀frozen24d ago29K32621184901.56%
OpenAI introduced ChatGPT Space, a new collaborative workspace for humans and AI Agents. > "In Space, you can work with pages—a new kind of interactive document with charts, images, checklists, and dashboards. > "Rolling out today to Pro, Business, and Enterprise plans." Pro testing time 👀frozen23d ago29K28319112601.29%
DAILY AI BRIEF 🗞 — Oct 15 SPACEXSI 🔥: - SpaceXSI launched Grok Voice 3.0, a speech-to-speech model built for customer calls that it says responds in under a second with 40% lower latency. - The Grok Bot marketplace is now open for developer submissions of connectors, plugins and MCP servers. ANTHROPIC 🔥: - Claude Code Desktop added a sessions hub for starting, tracking and following up on multiple sessions from one screen. - Claude Managed Agents, the Messages API and the Claude SDK now support OpenTelemetry traces. - Anthropic opened a Singapore office, its fifth in Asia-Pacific, and signed an AI safety MOU with Singapore's government. PERPLEXITY 🔥: - Perplexity Computer added Ultrafast versions of GPT-6 Astra and SOL 6.1 for Max subscribers. MISTRAL 🔥: - New Material, a design and AI product studio, is joining Mistral to help build its frontier solutions. MICROSOFT 🔥: - GitHub Copilot updated HydraFusion's model pool and added a new HydraFusion Max option, both still in research preview. GOOGLE 🔥: - Google Messages is adding an enhanced protection mode against phone scams. - YouTube Music is rolling out labels for AI-generated songs. OPENAI 🔥: - Codex got another global usage-limit reset on Day 10. NVIDIA 🔥: - NVIDIA AI teased its own model launch coming next week. * Used Grok to compose this brief, cherry-picking the news and doing some post-editing.frozen8d ago29K1285323190.64%
GOOGLE 🔥: It seems like the upcoming Gemini Task mode on Gemini desktop will support Gemini Live as well. > Users will likely be able to control Gemini Tasks by voice. > The Computer Use feature also got more granular settings, letting users configure whitelisted and blocked apps. These features already undergo closed testing, and we may see them announced soon. It would be cool if they arrive along with a rumored Gemini 4 Pro, which may drop earlier than expected this year.frozen28d ago29K46323175702.02%
GPT-6 Astra and GPT-6.1 Sol are now expected to be 50% faster. Halfultrafast?! 👀 Would you prefer this or banked reset? https://t.co/HLaEjblJjifrozen17d ago29K44815451241.87%
Google is working on "Superprojects" for Gemini. > Superprojects (internal name) are the next iteration of Projects, which will replace Gemini Notebook projects in Gemini. > Files and chat sessions in these projects will be accessible through various Google products. Soon? 👀frozen17d ago28K51736263882.37%
PERPLEXITY 🔥: A fully local option is now available on the Perplexity Computer macOS Desktop app for Pro, Max, and Enterprise subscribers! They did it 👀 Perplexity Computer no longer needs the internet or paid credits to work. A fully local mode is powered by PPLX Pro, a post-trained version of the Qwen 3.8 model. For complex tasks, it can also run in Hybrid mode to use cloud models. Perplexity Computer has a very good harness, and this needs a lot of testing time! * 36 GB of unified memory is recommended. https://t.co/rTuCE6k6Qa2d ago28K331163251241.82%
DAILY AI BRIEF 🗞 — Oct 8 ANTHROPIC 🔥: * Anthropic launched Claude Haiku 5.5 with a 1M-token context window and adjustable effort, priced at $0.10/$0.50 per million tokens. * Prompt cache reads on Claude Sonnet 5.5 now cost half as much, dropping from $0.20 to $0.10 per million tokens. OPENAI 🔥: * A new GPT-6 version with Intelligent UI is rolling out to all ChatGPT users, bringing interactive visual answers. * ChatGPT for Teens is getting a College Planner that tracks application and financial aid deadlines. MICROSOFT 🔥: * Copilot on Copilot+ PCs will be able to use your PC's files, take actions in Windows, and run local models, starting in the coming months. * Microsoft Execution Containers, a policy-based sandbox for AI agents on Windows, are now generally available. * Windows Search is getting thousands of quick actions from the taskbar, starting with Windows Insiders. * MAI Code 1.1 Flash is coming to local PCs in a 3-bit version that's nearly 80% smaller. * Surface Laptop Ultra with NVIDIA RTX Spark is up for pre-order from $2,599 and ships October 16. * GitHub Copilot's local sandboxing is now generally available across the CLI, the Copilot app, and VS Code. * GitHub Copilot CLI can now find and use local models from a running Ollama instance. SPACEXAI 🔥: * Grok Bot can now use Claude Opus 5.5, and Elon Musk says it will pick the best model for each task. * Grok Bot can now search X natively and analyze more than 100 posts from one prompt. * Grok Bot can now build slide decks and deliver them as PowerPoint files or Google Slides. * SpaceXAI is working on a Meetings feature that lets Grok Bot join Google Meet or Zoom calls. GOOGLE 🔥: * Google launched Playground, an experimental platform in the US for creating and sharing AI-generated games without code. * Google's SynthID Detector for checking AI-generated content is now available to everyone worldwide in English. * Antigravity is showing Google AI Ultra upgrade labels for Gemini 4 Argon, hinting at an upcoming rollout. * Google is prototyping a real-time voice agent in Antigravity, internally called Concierge. META 🔥: * Meta's Muse agent app is now available on iPad. * Muse is coming to Windows soon as a native app. NVIDIA 🔥: * NVIDIA said fine-tuned Nemotron models reached gold-medal level at both IOI 2026 and IMO 2026. PERPLEXITY 🔥: * Perplexity open-sourced pplx-embed-v2-late, multimodal retrieval models that search text, images and PDFs without OCR. LIQUID AI 🔥: * Liquid AI released d1-3B and d1-omni-600M, open multimodal decision models built for edge devices. CURSOR 🔥: * The Cursor iOS app can now see and reply to local agents running on your computer. * Used Grok to compose this brief, cherry-picking the news and doing some post-editing.frozen15d ago28K1529354190.79%
ANTHROPIC 🔥: Claude Sonnet 5.5 has been announced officially and it is 30% faster and 30% cheaper with a performance close to Opus 5.5! A new default 👀 https://t.co/CxiHQdYqQLfrozen24d ago27K42421141321.79%
OPENAI 🔥: GPT-6 Astra Pro got upgraded with a better use of plugins and tools along with a speed improvement. > Additionally, multiple agents can tackle different parts of a task in parallel! > This upgrade is now available to Pro users on Chat. Wondering if this will be expanded to Team accounts as well and if we are expected to see GPT-6.1 Astra Pro anytime soon. Would be neat 👀frozen6d ago27K49621253302.12%
Figure is planning to release its own proactive AI assistant this week! Here is what we know so far 👀 > Hark will likely run on Opus 5.5 for general tasks, with Gemini 3.1 Flash Image, Qwen3.6-27B, and GPT-6 Astra referenced in the code for other specific tasks. > Hark will arrive with Hark Pro, Pro², Pro³, and Pro⁴ paid plans, with 5× and 20× usage options and another offering twice the usage of Pro³. "Hark has the power to plan 800 weddings, buy 1000 grand pianos, move across the country 1,500 times, and hire 1,000 lawyers for when it all goes wrong." > Hark will have its own Memory and Soul, support for Projects, widgets, scheduled tasks, and many other features, including computer use. > Hark will be available on the web, iOS, and iPad; it will launch with a waitlist, and invite codes will likely be available too. Looks like Muse is about to get some competition, especially if Hark becomes available more broadly.frozen17d ago27K23417315811.37%
DAILY AI BRIEF 🗞 — Oct 20 OPENAI 🔥: - ChatGPT Slides now supports real-time collaborative editing, comments and on-slide edits for Pro, Business and Enterprise plans. - OpenAI Platform recently added a "Decisions" API playground for testing its decision model with Predicate, Choice, and Score modes. - An unannounced GPT-6.1 Terra model surfaced in a public Microsoft code update that switched 18 evaluation configs to it before being reverted. - Reuters reports OpenAI is speeding up a new fundraising round ahead of Anthropic's expected IPO. ANTHROPIC 🔥: - The Information reports Anthropic will release a new model this month that is especially good at reading users' underlying intentions. SPACEXSI 🔥: - Grok Bot got a limited-time Pumpkin Bot Halloween avatar, and iPhone users can now launch their Bots with the Action Button. - Runway MCP is now available in Grok Bot, bringing Runway generation into the chat. - X rolled out a new AI ranking model for timeline recommendations plus a retrieval model trained to surface posts people read to the end. META 🔥: - Meta published a technical report on Muse Glimmer, its 30B Apache 2.0 open-weight agentic model that runs on a single consumer GPU. MINIMAX 🔥: - MiniMax officially launched M3.1 Flash, the former "Space Bunny" stealth model, at $0.15/$0.60 per 1M tokens, with open weights due Oct 23. CLOUDFLARE 🔥: - Cloudflare introduced Clairic, an extraction model that pulls, computes, and forecasts data from unstructured inputs. INCEPTION 🔥: - Inception launched Mercury 3, a diffusion LLM that runs at over 1,000 tokens per second with a 1M-token context window. FAL 🔥: - fal and Black Forest Labs released FLUX 3 Max, a post-trained video model that generates a 10-second clip in about 12 seconds. * Used Grok to compose this brief, cherry-picking the news and doing some post-editing.3d ago27K1656264240.84%
BREAKING 🔥: OpenAI introduced "Ultrafast", a new premium speed tier. > "Ultrafast offers up to 8× faster token generation (300 tokens per second) in Codex and up to 6x in the API." > "GPT‑6 Astra Ultrafast is available today in the API⁠ and in ChatGPT Work and Codex⁠ on Pro 500 and Enterprise plans. GPT‑6.1 Sol Ultrafast is coming soon. Pro 500 is mentioned too 👀frozen23d ago27K33914191411.56%
ICYMI 👀: Codex can now use a reset on its own when users explicitly ask it to do so. > "Allow Codex to use resets - Give Codex the ability to use a reset when you explicitly request it during a conversation and your current limit has 10% or less remaining. Resets are never used automatically." This setting is placed in the Usage & billing section.frozen12d ago26K27121375321.39%
OpenAI seems to be preparing to release GPT-6.1 Terra model as well. > The model reference recently showed up in Microsoft's PR for evaluation configs. > Earlier, The Information reported that this model is also in the works. https://t.co/MbCQHsaOUf2d ago26K487255810362.39%
MICROSOFT 🔥: A new Copilot app with Home, Code and Copilot modes has been introduced! > “Today we’re introducing the new Copilot to connect the tools people rely on with the next generation of capabilities they’ll need to build, customize and scale AI across work.” > Home contains Chat and Cowork options for day-to-day work. Office Copilot is included. > Code is powered by the same underlying tech as GitHub Copilot. A new Microsoft Managed Copilot Runtime for code execution has been added too. > Autopilot is a rebranded Scout Agent, Microsoft’s proactive AI assistant. * “It’s rolling out now through the Frontier program” Frontier testing time 👀 https://t.co/lHZHwUMZUdfrozen27d ago25K33325277611.82%
Google is making Markdown files natively supported in Google Drive and Google Docs. Markdown files is already a default way to exchange context and instructions with AI agents. This change will simplify so much! https://t.co/jvGvjYNddXfrozen17d ago25K61040331823.09%
Inworld has acquired Ultravox, a platform for building voice agents that listen, reason, call tools, and respond in live conversations, handling turn-taking and interruptions along the way. Together, the two aim to bring speech understanding, reasoning, and expressive voice generation closer together for speech-to-speech agents. > Built-in Inworld voices on Ultravox now run on Realtime TTS-2. > Existing voice IDs keep working with no migration or code changes. > TTS-2 also adds natural-language steering to Ultravox agents.frozen22d ago25K38281160.26%
GOOGLE 🔥: A new Gemini Agent for Gemini Enterprise has been announced as a part of Gemini at Work updates. - Gemini, a single, universal agent for work that answers your questions, handles your knowledge work, creates your images and media, and writes and runs code.  - Inline in Workspace, Gemini works directly inside Gmail, Drive, Docs, Slides, Sheets, Chat, and Calendar, carrying the same memory, skills, and controls it has everywhere else. - New data and analytics skills that allow both technical teams and everyday business users to use plain-language questions to get to actionable operational insights in minutes.  - Industry-specific specialization with tools, skills, connectors and knowledge specific for financial services and legal teams.frozen14d ago24K41627244682.23%
Builders can now submit their Plugins and MCP Connectors via a new submission flow on Claude Directory. Plugin usage stats are available too! Plugins are new "apps" 👀 https://t.co/4i9sAN3L6Rfrozen27d ago24K2291270751.34%
ICYMI 👀: Google AI Studio Build now has Pan and Max modes available. > "Max mode - Work longer for better results" > "Plan mode - Create a plan before making changes" Gemini 4 Argon MAX 🔥 https://t.co/8noNcCb0TQfrozen6d ago24K41421362622.22%
BREAKING 🔥: Meta is rolling out Muse Wonder and Muse Wonder Full on the Meta Model API platform! > Muse Wonder Full - Our most capable model, running at max reasoning. For the hardest, longest-running work — available to Muse Code subscribers, not on the API. > Muse Wonder - State-of-the-art at real-world coding, built for long-running tasks. Plans the work, calls tools in parallel, hands parts off to sub-agents, and recovers from errors so complex jobs finish.frozen10d ago24K39624297542.13%
Gemini Notebook users will soon be able to generate videos and slides step by step from the chat conversation. A new hidden notice has been added recently: "You can now create Slides and Videos step-by-step" Gemini 4 upgrade soon? 👀 https://t.co/veR2VTD4qZfrozen6d ago24K48433273772.61%
DAILY AI BRIEF 🗞 — Oct 6 OPENAI 🔥: * GPT-6 Astra and GPT-6.1 Sol now run about 50% faster in ChatGPT, kicking off a 28-day daily-improvement pledge for Codex and Work users. * API customers can now opt in to textGrain text watermarking, with ChatGPT and Codex text in the EU getting invisible watermarks in the coming weeks. * ChatGPT will start testing visual ads during image generation in the US later this month. * The Wikimedia Foundation says rogue OpenAI agents edited its wikis and may have contributed to a partial outage in May. REFLECTION 🔥: * Reflection unveiled Beam, a 501B-parameter (23B active) open-weight MoE model, with Apache 2.0 weights due later this month. AMAZON 🔥: * Amazon Nova 2.5 Sonic is now generally available on Bedrock for real-time voice agents. * https://t.co/NwUG66ZSCF's GLM 5.3 is now generally available on Amazon Bedrock. COHERE 🔥: * Cohere launched North 2, its biggest platform upgrade yet, with a new agent harness and bring-your-own-model support. GOOGLE 🔥: * Google Docs and Drive now open, edit and render Markdown files natively. * Nano Banana 2.1 appears to be live in Google Flow. * Google is working on "Superprojects", the next version of Projects in Gemini, shared across Google products. * Gemini's Call for Me may expand from calling businesses to calling friends and family. * Google paused its open source bug bounty program until next year, citing a flood of AI-generated reports. ANTHROPIC 🔥: * Claude is now available with in-country inference in India through Amazon Bedrock. * Claude Code 2.1.290 adds claude attach and claude logs commands. MICROSOFT 🔥: * GitHub released ReviewBench, an open benchmark for AI code review agents. FIGURE 🔥: * Figure is preparing to launch Hark, its own proactive AI assistant, this week with a waitlist. HUGGING FACE 🔥: * Hugging Face profiles can now show your P(doom), feeding an anonymized survey on AI risk. * Used Grok to compose this brief, cherry-picking the news and doing some post-editing.frozen17d ago24K15711141210.85%
OpenAI updated ChatGPT Slides with real-time collaborative editing. > Teams, Dots, and ChatGPT can now work together on a slide deck in order to make edits, enhancements, and leave comments. The feature is available on ChatGPT Pro, Business, and Enterprise plans. https://t.co/MXmb0uRCGK3d ago24K45317321752.43%
Google seems to be rolling out support for custom MCPs on Gemini desktop app to more users. This means that along with Gemini 4 we may get additional updates - for example, Computer Use has been in testing for quite some time and should be ready by now. The same goes for “Superprojects”. Yet, the biggest win would be there if Google release Gemini 4 to Pro users as well. Is there a chance? 👀 h/t @BartokGabi17frozen9d ago24K43031453682.44%
GOOGLE 🔥: Gemini 4 Argon has started rolling out and is now available in Agent Studio on GCP. You can already test the model there 👀 Google AI Studio and Gemini apps should follow. Testing time! https://t.co/MndO9VVNmkfrozen9d ago24K44632273572.40%
DAILY AI BRIEF 🗞 — Oct 7 MISTRAL 🔥: * Mistral released Mistral Large 4 "Le Chonk", a 1T-parameter (49B active) multimodal model, with open weights planned in about three weeks. GOOGLE 🔥: * Nano Banana 2.1 is rolling out across the Gemini app, AI Mode, AI Studio, the Gemini API, Flow and Stitch. * Google released EmbeddingGemma 2, a 740M-parameter open multimodal embedding model under Apache 2.0. * Google AI Edge launched Foresight, an on-device note-taking app for Mac. OPENAI 🔥: * OpenAI launched the Decisions API in beta with gpt-6-luna, returning typed answers 10x faster than the Responses API. * Auto-review in Codex is now free for everyone signed in with ChatGPT, day two of the 28-day pledge. * OpenAI published new math results from an unreleased internal frontier model, with many proofs formalized in Lean. * OpenAI cut its API usage tiers from five to three: Build, Launch and Grow. ANTHROPIC 🔥: * Claude for Google Workspace is in public beta, adding a Claude sidebar to Docs, Sheets and Slides. * Anthropic's expanded Claude for Startups program gives qualifying startups a free year of Claude Team and $1,000 in API credits. * Claude Code 2.1.292 lets Claude run sub-agents at a chosen effort level. HARK 🔥: * Hark, Brett Adcock's new startup, launched Hark Pro widely as a free computer-use AI assistant, with a paid tier for heavy users. SESAME 🔥: * Sesame's voice assistant got a new voice model, a dedicated computer for each agent, and Google app and MCP connections. MICROSOFT 🔥: * GitHub stacked pull requests are now generally available. * Used Grok to compose this brief, cherry-picking the news and doing some post-editing.frozen16d ago23K1237262130.73%
OPENAI 🔥: Dots access will be included in Pro 100 plan and soon will be expanded further as well. How soon? 👀 https://t.co/pRdGocdGChfrozen23d ago23K2568143191.29%
BREAKING 🔥: OpenAI shipped an upgrade to its Voice Mode on ChatGPT, Codex and dots! A new model, GPT-Live-1.1, comes with support of new voices in French, German, Hindi, Mexican Spanish, Korean, and Japanese. A constant “Checking..” should be gone as well. It has been quite annoying during vibe coding sessions. API support is coming soon 👀 https://t.co/Ap8UKShiM6frozen7d ago23K43118465392.37%
I just realized that Grok Bots have been rolled out on Grok web and mobile! That's quite cool to have everything in one place. Wondering whether SpaceXSI will keep both mobile apps long-term. Also interesting to see how Grok Bot is slowly capturing the space where OpenAI Dots and Muse are not available.frozen6d ago22K33016388351.90%
GOOGLE 🔥: Nano Banana 2.1 appears to be available on Google Flow right now. Yet its output seems similar to what Nano Banana Pro generates, so it is unclear whether requests are actually routed to a new model. https://t.co/STtLs9xwOafrozen17d ago22K29616194391.69%
Mint is a new Green on ChatGPT app! The best change 💪👀 https://t.co/a9N1kx3X4jfrozen22d ago22K45213130412.36%
DAILY AI BRIEF 🗞 — Oct 21 OPENAI 🔥: - OpenAI acquired Orchestra Bio, a drug-development coordination platform, to build on its Rosalind life sciences models and tools. - Locked Chats are rolling out in ChatGPT across all plans, letting users protect chats with a PIN, Face ID, Touch ID or device unlock. - The OpenAI API now offers 12-hour prompt caching, so persistent agents can keep shared context cached at the lower rate for longer. - ChatGPT team tasks can now automate recurring engineering work, like a weekly roundup of what shipped and what's blocked. SPACEXSI 🔥: - SpaceX reported $14.6B in Q3 revenue, with its Super Intelligence segment (Cursor, Grok Bot and its neocloud) bringing in $7.5B. GOOGLE 🔥: - Gemini with Personal Intelligence is rolling out globally on Android Auto, with cars with Google built-in following in the coming months. - Google released llm-d 1.0, an open-source Kubernetes-native stack for distributed LLM inference, at the PyTorch Conference. MISTRAL 🔥: - Mistral released Voxtral 3, a family of speech-to-text and text-to-speech models powering its new voice agents API. PERPLEXITY 🔥: - Perplexity Computer on Mac can now run fully local, powered by PPLX Pro, a post-trained Qwen 3.8 model, for Pro, Max and Enterprise users. META 🔥: - Muse web is preparing Confidential VMs, an encrypted mode for running Muse tasks that will initially be limited to Meta AI subscribers. ELEVENLABS 🔥: - ElevenLabs released Eleven Realtime v1 Max and Mini speech-to-speech models on its ElevenAgents platform and APIs. * Used Grok to compose this brief, cherry-picking the news and doing some post-editing.2d ago22K1449182250.91%
OpenAI ❤️ Google Cloud GPT-6 Astra, GPT-6.1 Sol, and GPT-6 Luna models are now available on Google Cloud. > "Google Cloud is now the only hyperscale cloud provider offering access to all three leading closed-source LLMs" (aka, OpenAI, Anthropic, and Google). Goooood 😈 https://t.co/qS9z4L2SMh1d ago22K56423242663.10%
ICYMI 👀: Gemini Notebook got a new usage limits system, usage estimation for artifact generation, and a background queue for generations when you are out of quota. > Limits refresh every 5 hours - Instead of refreshing every 24 hours, your limits now refresh every 5 hours. Get continuous allocation throughout the day. > Create more of what you love - We are replacing limits per feature with a flexible Notebook-specific limit. You control how you want to spend your limits. > New background queue - When you are out of limits, you can defer generating videos, slides, etc. for later. These will generate automatically for you.frozen17d ago22K33520175461.96%
SPACEXAI 🔥: Grok Bot becomes proactive and can now suggest helpful things on its own. Users will be asked to configure a new Primary Bot or select an existing Bot as Primary (ex., your Chief of Stuff). The Primary Bot feature is rolling out gradually now! https://t.co/brtFG5ZB1wfrozen21d ago22K25011104351.44%
ANTHROPIC 🔥: Users can now modify Claude Code builds via Mods! > "Mods can rewrite or replace what Claude Code does, and can even draw custom UI. Under the hood, mods are hooks, and they ship inside plugins. Each one is a small JavaScript or TypeScript module that runs inside your session and sees every event as it happens." Mods can also be shared with others via Claude Directory. Tbh, you can now "mod" most of the client-side software with a bit of effort and tokens. These are first steps towards "personalized UIs," so to speak. https://t.co/Bqw9zLL26ffrozen21d ago21K2467133641.56%
Odyssey has released a playable research preview of Agora-2. Agora-2 is a new multi-agent world model that is able to generate multiplayer worlds in real time for up to 20 concurrent players and AI agents. > Every frame comes from the model, and each participant's actions change what everyone else sees. > Agora-2 supports up to 5x more participants than Agora-1 and can cover multiple environments at once.  > It can handle longer, larger, and more complex playable worlds.frozen28d ago21K1118102480.84%
SpaceXAI is working on Remote Control for Grok on web and Mobile. Interestingly, Remote Control feature on iOS is referring to a “Grok Desktop” app and not to Grok Build CLI. Considering that it’s a recent addition, does it mean that we will get a desktop app as well? Sus 👀frozen1mo ago21K2229175231.30%
ANTHROPIC 🔥: An early look at how agentic shopping flow with Stripe Link would work on Claude Duet, Anthropic's upcoming proactive agent. Interesting to see these hashtags under the agent - does it mean that we will see its UI designed around "threaded" conversations like on Slack? Fun fact: Duet's internal name is "aibo". APKPLZ 👀 https://t.co/H5FQREwhvl1d ago21K40315285712.52%
TinyFish has launched two free builder programs: One for students and the other open to anyone who wants to become an ambassador. > Both programs share the same monthly bounty task board. > Every submission that meets the brief can get approved. > Multiple users can get paid for solving the same bounty. Agentic Bounty Hunting 👀frozen24d ago21K137230971.15%
META 🔥: Muse Gadgets and Muse Home Link have been announced! > Muse Gadgets is an open-source ESP32 firmware and Linux SDK. > Muse Home Link is a device to allow Muse to work with smart devices. > Users can Claim their Muse Home Link device on the website! “Connect Muse to your home Wi-Fi so it can reach compatible devices you already own, or anything you build with a local HTTP API.” “Community-built skills can let Muse turn on a light, control your TV, send a document to a printer, and more, depending on your setup.” “Free with an active Muse subscription in the United States only, limit one per subscriber. Ships in October, first come, first served.” Cassandra testing time 👀frozen20d ago20K24111131441.52%
META 🔥: Muse Wonder Full scores second on the Artificial Analysis Index, just 2 points behind Claude Opus 5.5! > SOTA on DeepSWE, scoring 79 points! > Muse Wonder capabilities cover Active Memory, Active Focus, Frontier Coding, and more! > Muse Wonder is now available on the Meta Model API and in Muse Code. Muse Wonder Full is launching exclusively for Muse Code subscribers. "Muse Wonder and Muse Wonder Full apply specialized knowledge to problems people encounter at work and in daily life, from understanding a lease to making sense of financial or health information." Meta is among the top AI labs now 👀frozen10d ago20K40826327502.57%
OpenAI announced a new Meetings plugin for ChatGPT to let it take notes and save a summary in ChatGPT Space. > Available in beta for Pro and Business users in the ChatGPT desktop app on macOS. Enterprise is coming soon. Feels like I saw it somewhere before 👀 https://t.co/Pac11XKQO8frozen16d ago20K32216213482.02%
ChatGPT profiles now have new sections for Top Plugins and a Showcase with Sites. Users can configure the list of sites to showcase and make them publicly visible. Finally, something for non-Pro users 👀 Interesting to note that Perplexity was again leading the pack over here - we will see more and more "premium" experiences being the core focus for AI labs.frozen22d ago20K22913100391.45%
ICYMI 👀: Google is killing Gemini Gems on November 17, replacing them with Skills! It was originally planned for October 20, but it was delayed. https://t.co/HJNxVCLMhNfrozen22d ago20K36612165412.21%
SPACEXSI 🔥: A new unified X Subscription plan across Grok, Cursor, and X Premium is gradually rolling out! > Plus - $8/month > Premium - $30/month > Super - $100/month > Ultra - $200/month “Each plan on The X Subscription provides more usage-per-dollar of our own models than the comparable plan it replaces.” That’s an awesome change and a big win for SpaceXSI as well. For me, that’s a free Cursor unlock. https://t.co/GdTFYYgNGlfrozen10d ago20K26418352371.79%
OpenAI DevDay 2026 is starting soon 👀 Let's break it down 🧵 https://t.co/NYDV0G3B2cfrozen23d ago20K145570150.88%
SPACEXAI 🔥: Grok Bot can now search X! Earlier, Grok Bot would have to rely on web search or the X connector; with this release, it can match Grok's capabilities, where one prompt can trigger analysis of more than 100 X posts. I use Grok Bot on a daily basis to compose a Daily AI Brief - looks like it will get much better tomorrow. Testing time! 👀 https://t.co/IzgVsKrvh4frozen15d ago20K32625134552.16%
ATOMIC 🔥: A new open-source Atomic Agent Desktop app is now available for macOS, Windows, and Linux! The app lets users run open models like Qwen and Gemma locally. The app is free, requires no account, and connects to a cloud model only when selected. > A setup wizard checks the machine's RAM and suggests which models will run well. > Fusion lets a cloud model plan a task while up to 8 local agents carry it out. > You can import setups from Claude Code, Codex, Hermes, and OpenClaw in a few clicks.frozen14d ago19K304241922373.09%
GOOGLE 🔥: Gemini 3.8 Flash TTS and Gemini 3.8 Flash-Lite TTS are now live on Google AI Studio and APIs! > "Flagship TTS model for Voice Design and dual-speaker screenplay control. Prompt custom vocal personas, direct line-by-line delivery, and add vocal bursts." > "Our most expressive audio generation models yet. TTeSting time! 👀frozen29d ago19K22215152341.54%
SPACEXAI 🔥: Grok 4.7 is rolling out on Grok web and mobile apps! It is now a base model across all modes, including Fast, Expert, Build, and Heavy. Looks like DAILY AI BRIEFS are about to get better! We'll see tomorrow 👀 https://t.co/V4rQ63d3UEfrozen21d ago19K288841171.71%
OPENAI 🔥: "10am PT, on the dot", OpenAI DevDay keynote is set. "Dots" are coming! https://t.co/41TGjEZtxH23d ago18K2949183141.85%
META 🔥: Confidential VMs feature is in the works on Muse web. Meta previously teased that this feature will go live before the end of 2026 and will let users run Muse tasks in an encrypted, confidential environment. According to a new hidden setting, Confidential VMs will initially be available only to Meta AI subscribers. Voice mode, image and video generation, and wearables support won't be initially available if the user switches to Confidential VMs. > "It stays between you and your Muse - Your conversations run in a secure, encrypted environment." > "Your Muse data won't be used to improve AI at Meta." > "Some features aren't available yet - This includes voice mode, image and video generation, and wearables." > "Availability is limited. For now, Confidential VMs are only available to subscribers."2d ago18K2389536211.79%
ANTHROPIC 🔥: Claude Duet, an upcoming proactive agent from Anthropic, will support a wide range of connectors, agentic shopping, and secure login. > "Claude Duet can purchase things for your business using Link, Stripe’s digital wallet." > "It can also use Shopify's Shop Pay to check out at online stores." > "Coming soon, it will be able to sign in to sites with logins you’ve saved in 1Password."1d ago18K40325251492.77%
Meta is working on “Agent Engine” for its Meta API Platform. Agent Engine is called “Forge” internally, however, it is not yet clear what solution will it be offering. A new harness model incoming? 👀 https://t.co/6RukhrZb0kfrozen19d ago18K1978256271.46%
Meta Connect keynote is happening soon, and there are high expectations for this event! What do we expect? 🔮 - A new Watermelon class model. - Muse Spark open source. - Muse Video launch and API availability. - New voice models and voice creation features. - AI for banking and finance and Oracle Cloud integration. Muse updates? 👀 - Connectors interface for Muse on the Meta Platform. - Shared Muse agents. - Telegram and Messenger support for Muse. - Muse for Meta Glasses. - Muse Mail. - New shopping and payment features for Muse. - Muse integration with Meta Business tools. - Custom voices for Muse. And obviously a lot more! What are your hopes? Event live thread 🧵frozen29d ago18K1887252321.41%
OpenAI published a repository with novel solutions to a range of math problems solved by their next frontier model. > The vast majority of results were obtained with the same procedure using an unreleased internal OpenAI model. > On average, each result used three hours of ChatGPT Pro thinking compute with that model. > Over the course of the evaluation, the model was posed approximately 4,000 problems.frozen16d ago18K24615151231.67%
OPENAI 🔥: GPT-6 and Intelligent UI are rolling out to all ChatGPT users. Intelligent UI lets ChatGPT create an interactive experience to explain requested topics. It seems like we are not getting GPT-6.1 on ChatGPT yet. https://t.co/KkbighJAtWfrozen15d ago18K4251890492.82%
Anthropic expanded dynamic workflows in Claude Managed Agents into a public beta. > Users can now configure their agents for multiagent orchestration, letting Claude plan and operate a fleet of agents to achieve the goal. Agentic loops invoking multiagent orchestration workflows will be insane. A new SWE norm 👀 https://t.co/gxGZZjFK61frozen13d ago18K28018367662.30%
Skyfall AI released SKY-001, the first World Model specialized in engineering tasks. > According to Skyfall tests, SKY-001 outperforms GPT-6 Astra and Opus 5.5 on CAD and BIM tasks. > The model sets a new bar for speed and cost per sample, running about 120 times faster and costing about 185 times less per case. “SKY-001 is currently a Compositional Visual World Model, aka designed to provide a visual plan of an engineered object.“ This model breaks physical objects into small parts and rebuilds the whole thing in CAD, as if it were a real factory!3d ago18K21816154841.91%
BREAKING 🔥: Anthropic is rolling out Claude Sonnet 5.5 on Claude and APIs! Testing time! 👀 https://t.co/xBedPdtQg3frozen24d ago17K259990231.72%
New Meta VR Glasses is a tech from the future 👀 https://t.co/Z8f4DY7FYvfrozen29d ago17K186551161.22%
Nace AI has released Drex Max, its most powerful decision model to date, which can now read images, videos, and text. Computer Use agents can send Drex Max a screenshot at each step, ask which action to take from a fixed set of options, and receive a probability score for each one. > Drex Max doesn't write text; it replies only with one of the options it's given. > Web agents can use it to select the right element of a page based on what is on screen. > It runs up to 167× cheaper per token than the models behind Claude, GPT, and Gemini for computer use.3d ago17K22510280932.06%
GOOGLE 🔥: A new Playground experiment is now available in the US. Users on paid plans can generate, while everyone can play games available in the gallery. https://t.co/eA87VMMjltfrozen15d ago17K25415142572.00%
ANTHROPIC 🔥: Claude is now available in Google Docs, Sheets, and Slides! Besides that, these file types now open inside Claude as well. A small change but a big quality-of-life improvement! https://t.co/IttFSHz73Cfrozen16d ago17K30717203472.32%
Google is working on Security review featue for Google AI Studio. > "Apply secure coding practices". > The new mode may appear in the mode selector, along with a Plan mode (also in development). Will it be powered by Gemini 4rgon? 👀 https://t.co/7ZpsvhLPHqfrozen22d ago17K2439180241.75%
GOOGLE 🔥: Nano Banana 2.1 is rolling out on Google AI Studio, APIs, and Gemini! > "High-quality image generation and editing model with enhanced factuality, visual fidelity, and multi-turn consistency at Flash scale." > Image output is priced at $30 per 1,000,000 tokens. > Output images at 1K (1024x1024px) consume 1120 tokens and cost $0.034 per image. > Output images at 2K (2048x2048px) consume 1680 tokens and cost $0.050 per image. > Output images at 4K (4096x4096px) consume 2520 tokens and cost $0.076 per image. Based on the pricing, it looks like it is a Pro model!?frozen16d ago17K28717321352.24%
Looks like profile picture support is coming to Claude apps! This feature has been in development for a while. Did you get it too? https://t.co/g4b3mhOkJFfrozen20d ago16K2081160141.46%
ANTHROPIC 🔥: Claude Dashboards is now available on all paid plans; Claude Motion is now available on Team and Enterprise plans. > Claude can query your data platform or CRM tool to build interactive dashboards or short animations. Everything is generated as code 👀 Another proof that Anthropic's focus on software development was the right direction, and it allowed them to slide into the enterprise world very seamlessly. That's a huge deal for analytics teams and product manager roles, as well as thousands of others. https://t.co/0hGuuytkSgfrozen14d ago16K25920221532.18%
Meta has announced the Meta Enterprise Platform to expand its AI stack to organizations and businesses. > "Meta Enterprise Platform will focus on turning its AI stack into products and services that companies can deploy for their own businesses. > "As with Muse, security and privacy are built into Meta’s enterprise products from the outset. Earlier, we also reported that Meta is preparing a strategic partnership with Oracle. We may hear more about that soon. Metaprise 👀frozen24d ago16K1697121151.29%
Wajo has launched Fo, an AI agent that can place calls, send emails, and pay with its own card to complete everyday errands. > Fo can send emails from its own address and check out on its own using a single-use card for each purchase. > Handle the back-and-forth with vendors over calls, emails, and group chats. > When a task requires a person, Fo is able to hand it over to one of its trained executive assistants. In Wajo's tests on tasks with setbacks, such as sold-out slots or unresponsive vendors, Fo completed 71% of tasks without help and stayed within user-approved limits on 94% of test runs.23d ago16K872121280.82%
BREAKING 🔥: Google has officially announced Gemini 4 Argon to be available on the APIs and on Gemini for Ultra subscribers. > "For developers and enterprises, it’s available in the paid tier of the Gemini API and the Gemini platform, for Ultra subscribers in Google Antigravity, and Ultra and Pro subscribers in the Gemini app and Search." Ultra testing time! 👀 https://t.co/UsGK0qeqDPfrozen9d ago15K26013202312.11%
Grok Bot has a limited-time Pumpkin Bot Halloween avatar, and iPhone users can now activate their Bots via the Action Button. I am slowly moving more and more of my routine tasks to Grok Bot. For EU users, it is one of the top solutions atm. Besides Pumpkin, we need a Ghost Bot too!3d ago15K20412122201.62%
OPENAI 🔥: GPT-6.1 Sol Ultrafast is rolling out on ChatGPT Work, Codex, and the API. > GPT-6.1 Sol in Ultrafast mode is $12 per million input tokens and $60 per million output tokens. > Ultrafast mode delivers 8x faster speeds than Sol Standard. Ultrafast testing time 👀 https://t.co/XgtmJefup4frozen14d ago15K22815262191.88%
SESAME 🔥: A huge upgrade arrived for the Sesame voice assistant, along with an updated, more expressive model, image generation, connectors, computer use, and more! Sesame can now operate coding agents like Devin, Codex, or Claude Code. ALL IN VOICE! 👀👀👀 https://t.co/43ZD7K1rR6frozen16d ago15K21317253652.09%
Vellum has launched Ambient Assistants, adding @vellum_ai assistant as a floating pill on top of any Mac app, where it sees the screen in real time and can draw on top of it to point at what matters. Context that used to be taken from screenshots, pasted links, and long prompts now comes straight from whatever is open on screen. > Users can hand off a task by voice without leaving the app they use, while Vellum assistant works on it in the background. > In unfamiliar tools, Vellum assistant can circle the exact button, setting, or metric to use, directly on screen. > Users can show a routine on screen and talk it through, so the assistant can run it later after its plan is approved.frozen28d ago15K54260280.59%
SPACEXAI 🔥: Grok Bot users now can ask their bots to claim its own email address. This address can be used by the bot to communicate with others, sign up for newsletters and more! Just asked it to subscribe for TestingCatalog’s daily AI Brief newsletter and it handled it perfectly. Testing time! 🤖frozen13d ago15K2379175332.00%
ICYMI 👀: OpenAI Platform now has a "Decisions" API playground. Users can test OpenAI's decision model with the "Predicate," "Choice," and "Score" options, which were released recently. Probability True 🤖 https://t.co/zpg2obCDdS3d ago15K18573010331.77%
NVIDIA 🔥: A new Nemotron 3.5 Super VL model has been announced! > NVIDIA Nemotron 3.5 Super VL is an open model with 120B total parameters and 12B active. > The model supports reasoning, tool use with text, image, document, and video understanding for long-running AI agents. https://t.co/9u0dzq19j17d ago15K23217313422.21%
SpaceXAI is working on a "Meetings" feature for Grok Bot. > "Give your Bot links to join Google Meet or Zoom" Imagine having a daily standup with your Grok Bots every morning. The future 👀 https://t.co/EniobkyPJNfrozen15d ago15K29614248382.61%
Hermes desktop users now can watch bot’s screens while they work and take the control over when needed. “Watching how someone else works is one of the most joyful states for a human being” I need to teach them “testing” 👀 https://t.co/SiKpioTdKYfrozen28d ago15K1799151541.78%
MICROSOFT 🔥: MAI-Transcribe-2-Streaming, MAI-Voice-2.1 and MAI-Voice-2.1-Flash models are coming to MAI Playground and APIs! > "MAI-Transcribe and MAI-Voice models: accurate, fast, low cost, and chart-topping audio understanding and generation for building the best conversational voice agents." MAI Playground has a new "Chatter" demo voice experience as well. Testing time 👀frozen21d ago14K166980341.50%
MICROSOFT 🔥: Copilot for Windows is getting Hybrid intelligence through local context from your Windows PC and local models that power Copilot tasks. > "With intelligent routing, powerful local models, and a performant runtime, Windows lets work run in the right place for each task — local or cloud. We’re bringing frontier-class capabilities to local PCs and enabling services like GitHub to route intelligently on Windows." > "With your permission, Copilot will be able to use local context from your PC, take action on your behalf, and tap into local models, bringing you more capability while helping your tokens go further." Windows testing time! 👀 https://t.co/Xrvmaoy5cPfrozen15d ago14K19112142401.81%
MICROSOFT 🔥: MAI-Base-2 pre-training has been completed; MAI-Base-3 training cycle is already being prepared as well. > "MAI-Base-2 outperforms every open-weight model we tested on our held-out evaluations. It is 4-12x more pre-training FLOPs-efficient than the strongest open base model, DeepSeek-V4-Pro-Base (see Figure 3 below), and we have put that saving into training a larger model." * Figure 1 shows the performance of MAI-Base-2 and other models with at least one trillion total parameters across a range of pre-training evaluations (lower is better).frozen7d ago14K20015187391.96%
MICROSOFT 🔥: A new Microsoft-Decision-1 model for fast decision-making is now available on Microsoft Foundry. Microsoft-Decision-1 was post-trained on Qwen3.5-9B for fast, single-pass decision scoring. Later, Microsoft is planning to rebase it on other models, including MAI and OpenAI models. > "Microsoft-Decision-1 achieved the highest accuracy in our 36-benchmark comparison, spanning nearly 150,000 questions across benchmarks kept blind from training." > "4.5 times quicker than Quyet-1.0-Large, the runner-up, and 35 times quicker than GPT-6 Sol." > "We’re already testing it across Microsoft for everything from incident response and quality control to scientific discovery." Everyone is testing 💪👀 https://t.co/ZI28j79Q0nfrozen13d ago14K1639246471.75%
Mistral Large 4 scores 62% on DeepSWE, outperforming GLM-5.3, according to VentureBeat. Additionally, it scores 67% on Finch (Financial tasks, SOTA open-weight). Le Chonk also scores 15% on Harvey’s Legal Agent Benchmark (Legal tasks, SOTA open-weight). We need a tech report now 👀 h/t @AiBattle_frozen16d ago14K22112122191.88%
InstaCloud has launched an agent-native serverless cloud, letting coding agents like Claude Code and Codex deploy and run the apps they build. It allows agents to handle provisioning, scaling, and monitoring, with users approving critical changes. > Services can scale with traffic and drop to zero when idle. > Agents can create branches of a whole environment, data included, to test changes safely. > The same setup can host apps, agents, and tools like n8n or Grafana.23d ago14K611100320.74%
ElevenLabs launched Eleven v4 and Eleven v4 Turbo, with Eleven v4 taking a first spot on the Artificial Analysis Provider Voice TTS Arena. > "Eleven v4 is built on a new architecture to generate speech with the intended tone, pacing, emotion, and character." A new TTSOTA 👀 https://t.co/Ssn12qjUxTfrozen24d ago14K150881331.44%
Minimax is preparing to release MiniMax-M3.1-Flash, as outlined in the API reference docs. > "MiniMax-M3.1-Flash - Frontier multimodal coding model with 1M context window and tunable thinking depth." > "Tunable thinking depth" seems to be a new perk the 3.1 family will offer. Soon 👀frozen3d ago14K24011382112.18%
Maket has launched Maket 2.0, an all-in-one AI platform for designing a new home or reworking an existing one. Maket 2.0 lets users request an extra bedroom or a larger kitchen and quickly get a floor plan and 3D view before any architectural work begins. > Floor plans can be generated from a short prompt, with an assistant asking about floors and rooms. > Existing layouts can be uploaded as images or PDFs and edited in chat. > Each room can be rendered in a wide variety of interior styles. https://t.co/44kBjvOcuUfrozen20d ago14K130793831.68%
Minimax officially announced MiniMax M3.1 Flash, formerly known as "Space Bunny" on OpenRouter. MiniMax M3.1 Flash scores 74% on DeepSWE and costs $ 0.15/$0.60 per 1M input/output tokens. Weights are coming on October 23. In addition, Minimax is preparing to release a Pro model too.frozen3d ago14K19813336332.05%
Sesame has announced its own eyewear product, launching in 2027! > "Our collection will be Made in Japan, available in 2027." As we reported earlier, Sesame is working on new payment plans, browser use support, scheduled tasks, connectors, and loads of agentic features for its AI assistant app. One of the most exciting products for me to monitor, besides major labs. Soon 👀 https://t.co/UuEn6nnSsLfrozen16d ago13K112643201.09%
Skills are coming as a separate attachment menu for Claude chats on mobile. That makes it way simpler 👀 https://t.co/yIywQxCKlNfrozen22d ago13K208871281.90%
Wabi has launched its AI messenger app to the public, with an AI assistant that users can add to group chats with friends and family. Wabi can join a group chat and monitor the conversation as people talk, offer help, and act when needed, like updating a shared calendar when something gets rescheduled. > Each plan can be split into threads, where Wabi can create a custom tool, such as a map of restaurant picks. > Wabi can connect to Gmail, Google Calendar, and other accounts, so it can automatically spot a problem and start a conversation about it on its own. > It waits for user permission for any important actions, and users can remove connections at any time.3d ago13K952232301.16%
Almanac Health has launched an AI workspace where physicians can hand off clinical and medical research tasks to AI and return when the results are in. > It can explore medical research, run clinical calculators, find relevant clinical trials, and take notes during patient visits. > Almanac is able to request clarification and ask for permission before taking an action. > Physicians can connect existing tools, including EHRs, and organize conversations by patient.frozen29d ago13K65040260.73%
Meta released many connectors for Muse to support new use cases for small businesses. > Asana, Box, Canva, Dropbox, Figma, Granola, HighLevel, Intuit QuickBooks, Klaviyo, Lovable, Notion, Shopify, Slack, Stripe, Zoom, as well as your Facebook and Instagram business accounts, are now supported. Muse is probably one of the most awaited products globally at this moment. Soon? 👀 https://t.co/aE048NP6Ob23d ago13K1481062281.48%
NVIDIA 🔥: A new model is set to arrive next week, possibly at the upcoming NVIDIA GTC event on Wednesday the 21st. Are we getting the recently spotted Nemotron 3.5 Super VL, or something else on top of it? > "Nemotron 3.5 Super VL combines a Nemotron-H hybrid language decoder with a RADIO vision encoder, a vision projector, and a separate temporal video embedder. It reuses the Nemotron Omni image/video stack without an audio encoder and retains shared Multi-Token Prediction (MTP) weights." Monitoring 👀frozen8d ago13K19212343322.09%
OpenAI ❤️ Orchestra Bio > OpenAI has acquired Orchestra Bio⁠ to work on Rosalind models and tools for life sciences research. > "Orchestra Bio is a platform that helps biopharma teams coordinate the complex process of drug development." AI advancements in life sciences are probably one of the most desired areas for many. So far, it seems like a top priority for Anthropic, OpenAI, and Google.2d ago13K27619121372.65%
Mitra has launched a personal AI assistant that makes and answers calls from your own phone number. You can listen live, whisper instructions, or take over at any point. Besides calls, @mitra_ai can send texts, emails, and Slack messages from your accounts, book appointments, and handle tasks on the web. It remembers context and follows up proactively 🔥 Available now on iOS, iPad, and desktop! https://t.co/XRoTFquFSqfrozen29d ago13K1041181581.41%
Claude Managed Agents, the Messages API, and the Claude SDK now support OpenTelemetry traces. > By default, they include duration, status, token counts, model, and tool names. Anthropic is steadily enhancing its cloud platform, and supporting OpenTelemetry is a major step toward enabling AI observability with tools like Grafana, Splunk, and others. Traces are useful for testing too! 👀 https://t.co/hi444jZOyqfrozen8d ago13K13010255261.53%
SPACEXSI 🔥: Grok Bot now supports live artifacts where it can build reports, memos, and dashboards. Users can ask Grok Bot to edit these pages or keep them updated over time. https://t.co/6gqJt91cut22h ago13K19412133342.02%
Base44 has introduced Base Code, its new cloud code editor that can open any GitHub repository directly in the browser. It works with multiple AI models side by side and lets team members with access to a repo turn a query into a pull request. > Base Code can handle existing codebases, including apps built outside Base44. > Developers can still review and approve every change in the PR before merging.frozen24d ago13K53170140.60%
Odyssey has launched Odyssey-3, its most powerful foundation world model to date, and opened a research preview to the public. Odyssey-3 builds an environment from a prompt and predicts changes in real time as a person moves through it or triggers events. > Odyssey-3 Pro scored 66.1 on Physics-IQ Verified video-to-video with best-of-8 sampling. This is the highest reported result. > Users can walk through scenes in first or third person. The camera can also move independently. > Odyssey adapted the model to robot arms, humanoids, and a car on real roads in India. Each system uses its own trained policy.frozen14d ago12K134690391.52%
Black Forest Labs announced FLUX 3 Image, a new image model for precise editing with 4K support. Users can already try the new features in the Tools Playground. Open weights are planned for release in the coming weeks. https://t.co/5Wp59Et1kpfrozen21d ago12K116242271.22%
DAILY AI BRIEF 🗞 — Oct 22 BREAKING 🔥: - PHP 8.5.12, 8.4.27, 8.3.36 and 8.2.35 fix two critical unauthenticated remote code execution flaws dubbed php2shell (CVE-2026-109618 and CVE-2026-109619), which OpenAI's Patch the Planet initiative helped uncover and now offers a Codex skill to check affected sites. - Microsoft Threat Intelligence reported that China state-aligned threat actor Storm-0940 used the agentic AI penetration testing framework ARTEX to automate attacks on hundreds of organizations worldwide. - Cloudflare warned that private keys for eight widely trusted eMudhra intermediate certificates were compromised, letting attackers forge certificates for any domain. - Node.js shipped security releases for its 22.x, 24.x and 26.x lines fixing multiple vulnerabilities, including high-severity issues. - Cloudflare warned that CVE-2026-88771 lets attackers run commands on NetScaler Gateway before anyone signs in. ANTHROPIC 🔥: - Anthropic opened a waitlist for Claude Duet, a standalone proactive agent app built around Claude Opus 5.5 that can also shop via Stripe Link and Shop Pay, for Pro and Max users in the US. - Claude Security now re-scans connected repositories as code changes land, in public beta on Claude Enterprise. - Anthropic released API version 2026-10-21 with new SDKs and ant CLI 2.0, now licensed under Apache 2.0. OPENAI 🔥: - GPT-6 Astra, GPT-6.1 Sol and GPT-6 Luna plus Codex workflows are rolling out in private preview on Google Cloud's Model Garden. - ChatGPT Ads expanded to 26 new countries, with more than 100,000 advertisers on the platform so far. SPACEXSI 🔥: - Grok Bots are rolling out to X Chat, letting you message your bots like any contact without leaving the X app. - Grok Bot can now build reports, memos and dashboards as live pages you can edit together or have it keep up to date. GOOGLE 🔥: - Gemini is now available in Google Maps navigation on CarPlay for parking, traffic reports and route changes by voice. META 🔥: - Meta reported $64.6B in Q3 revenue, up 26%, with Zuckerberg saying more than 25% of Muse users have bought items through the agent. NVIDIA 🔥: - At GTC Berlin, NVIDIA introduced the OpenShell Reference Architecture for securing enterprise agents and DSX POD, cutting AI factory deployment at prepared sites to as little as 14 days. - NVIDIA released Nemotron 3.5 Embed, Rerank and Parse open retrieval models on Hugging Face. MISTRAL 🔥: - Mistral partnered with NVIDIA on AI for industrial engineering, covering faster simulation, open physics-AI benchmarks and agentic evals for CAE and EDA. KLING 🔥: - The Kling 4.0 API is now officially open, with the model also landing on Runway, fal, Higgsfield and Adobe Firefly. COHERE 🔥: - Cohere launched Rerank 5 in Pro and Fast tiers. RECURSIVE 🔥: - Recursive released Flint, a 40B MoE open-weight local coding model under Apache 2.0 that it says is up to 2x faster than comparable models. * Used Grok to compose this brief, cherry-picking the news and doing some post-editing.13h ago12K1342322201.58%
Kled launched V3. Custom AI data collection tasks can now be deployed to 500,000+ contributors within 72 hours. 108 configurable task templates across image, video, audio, text and annotation. Everything from egocentric video to conversational speech in rare languages. 8 million datapoints collected per day. The real world is becoming programmable 🌍frozen22d ago12K73340200.83%
ICYMI 👀: ElevenLabs released Eleven Realtime Max and Mini speech-to-speech models. > Eleven Realtime v1 Max scores 98.8% on Big Bench Audio and starts speaking in 0.57s. > Eleven Realtime v1 mini is optimized for natural conversations at low cost. Both models are available on ElevenAgents platform and APIs. https://t.co/NIOfcCagfd2d ago12K14713335341.95%
OpenAI is planning to raise another $30 billion at around a $1.4 trillion valuation (not including the money raised), according to Bloomberg. Whose IPO will be bigger? 👀 https://t.co/sGlzr128Oefrozen23d ago12K202751151.95%
Krea AI introduced Krea 3, a new image model designed for fast, precise image editing. > Krea 3 combines separate references for characters, their outfits, and the location into a single scene while keeping faces and color grading consistent. > On CPI-Bench, a benchmark for real-world image editing, Krea 3 Ultra scored 4.81 out of 5, second only to GPT Image 2.5 and ahead of every other image model. Testing time! 🔥 https://t.co/3p6SRKIu7vfrozen7d ago12K1568113471.93%
Imagine testing this for the 1st time 🥵 https://t.co/Rul6L5FAXJfrozen7d ago11K2646131132.62%
Synthesia has launched Sessions, a set of interactive AI avatars for live coaching, interviews, and guided conversations. > Roleplay Sessions are designed for practicing sales calls and job interviews with an AI avatar in real time. > Survey Sessions use an avatar to conduct interviews and collect feedback or insights. > After each roleplay, an AI coach scores every skill and walks the learner through the results. Where Synthesia’s avatars previously delivered one-way training videos, they can now hold conversations and respond in real time!frozen21d ago11K62391120.77%
Tavus has introduced Griffin, a research preview of its newest video-to-video model for real-time face-to-face conversations. Griffin is designed to learn conversational behavior from people. It can watch and listen during conversations, allowing it to interrupt mid-sentence or respond when someone enters the frame. > Tavus says Griffin is the first model to pass the Turing test, with 45% of 120 study participants taking its conversation as real. > On NVIDIA's Video Full-Duplex Benchmark, Griffin Lite ranks first among every model evaluated, at 3.83 out of 5 against 3.92 for real humans. > The full Griffin model remains unreleased while Tavus reviews its release policy.frozen21d ago11K91480291.17%
Thesys has released OpenUI 1.0, its open-source framework that lets agents create working interfaces inside apps. On Thesys' own benchmark of 46 interface briefs across six models, OpenUI used 58% fewer output tokens than Vercel's json-render and 52% fewer than Google's A2UI. > OpenUI allows models to write a compact UI language that renders on screen while the answer is still being generated. > It works with any model, including open-weight ones running on a team's own hardware. > Version 1.0 adds custom actions, allowing buttons inside a generated answer to activate an app's own code.frozen7d ago11K11310151812.02%
NVIDIA GTC Berlin Keynote is about to start soon! GPUs, Nemotron, Robotics and loads of other updates. Gonna do some live text reporting here. Watching 🍿 https://t.co/7jzeErwuRM2d ago11K64516260.85%
Creatify has launched Boreal-H3, a video model for ad production, and an Ad Agent powered by Claude Opus 5.5. Boreal-H3 is post-trained on MiniMax H3 with real ad projects. Creatify measured brief success rates rising from 28% to 50%, identity match from 83% to 94%, and a 71% drop in failure severity on the same tasks. > Products keep their shape, labels, and details as the camera moves around. > Generated creators can handle UGC, complex spoken lines, and two-person dialogue. > Camera movements and keyframes can be directed by users, including for product handling and physics. The sample ad below was produced by Ad Agent in a single run.frozen22d ago11K76230301.04%
A small model drop from Liquid AI 👀 Liquid AI has released a new open-weight LFM2.5-24B-A2B model with 24B total parameters and 2B active parameters. Based on benchmarks, LFM2.5-24B-A2B performs on par with the larger Qwen3.6-35B-A3B model. The model is available on HF 🔥 https://t.co/ssrKPLq3hVfrozen9d ago11K1209223201.63%
Ideogram released Ideogram 4.5 image model for precise editing! > The model is available on Ideogram and APIs, as well as partner services, and the team is planning to publish open weights of this model soon as well. > The price ranges from 0.8 to 22 cents per image at native 2K resolution. Testing time! 👀 https://t.co/V4e2v8P0BCfrozen22d ago11K1941170572.53%
SPACEXSI 🔥: Grok Voice 3.0 has been announced, and it is now a SOTA speech-to-speech model on many benchmarks and the Artificial Analysis Speech-to-Speech Index. Grok Voice 3.0 also comes with 40% lower latency. AI call center testing time 👀 https://t.co/clyoAsBTJIfrozen8d ago10K25017210222.97%
Mitra AI assistants can now communicate for you in messaging apps and talk to other Mitra assistants via Mitra Link! > Users can mention their Mitra agent in the chat to delegate the ongoing conversation. > Via Mitra Link, users can choose exactly who can talk to their Mitra agent, enabling loads of new potential workflows. > Mitra assistant can handle incoming requests to align on the meeting time, scheduled events, and more! Mitra is designed as a personal AI assistant that operates through a user’s existing phone number. It can answer missed calls, make outbound calls, navigate phone trees, handle hold times, filter spam, and escalate conversations that require the user's direct involvement. This is quite a unique offering 👀 https://t.co/SRFh5Bq1DYfrozen16d ago10K73342210.99%
Paper Instruments has released Ultramarine 1 Flash, its new open model made for document-heavy workloads of banks, law firms, and consultancies. The company states the model can deliver Excel models, slide decks, or Word memos for cents per task. > The model is a post-training of GLM-5.3 Flash. > On Paper Instruments' internal benchmark for client-ready deliverables, it scored 59.3 at $0.13 per task, versus 57.8 at $8.19 for Claude Fable 5. > Each training task was graded on its own rubric, since a diligence memo or a board deck has no single correct answer.frozen8d ago10K94883341.43%
Elorian has released Glimpse, its first AI model, which can go back and reanalyze part of an image or video mid-thought. Glimpse can return to a specific time in a video while it thinks, zoom in on a detail, or replay half a second of video at a higher frame rate before answering. > On Elorian's own tests, Glimpse scored better than GPT-6 Astra on scientific chart reading and short-form video captioning. > For planning a simulation of a robot arm placing cream cheese in a bowl, Glimpse cost $0.035 vs. $0.249 for GPT-6 Astra. > In the demo below, Glimpse solved 5 mazes of increasing difficulty, while Gemini 3.8 Flash solved 1/5.frozen6d ago10K826182191.24%
Nature Will @naturewill has launched Natura-1, its first foundation model for the physical world! Natura-1 is trained on one million hours of egocentric video data of humans handling different scenarios and objects, enabling it to handle objects and tasks it has never seen before. > The model allows instructions to get specific, down to where an object should end up and which way it should face. > It can maintain control as its grip shifts mid-task, and do different tasks with each hand. > On longer tasks, Natura-1 can track what it has already moved or opened and determine the next step. https://t.co/Fj3q6i2OGqfrozen9d ago10K774194171.19%
Mistral AI ❤️ New Material New Material is joining Mistral to help build its frontier solutions. > "New Material have shipped award-winning AI-native tools, products and ventures across automotive, retail, legal and healthcare with partners such as Ford, Rafael Nadal, Wellcome Trust and AmLaw 50 firms, among others." Le New Material 👀frozen8d ago10K1391080141.69%
Glasser released a pay-per-call data service for AI agents! It allows agents to access ~2000 paid API endpoints across different providers, under a single key without a subscription. Users can get started by typing "set up https://t.co/WCMUOBdw5M" into Claude Code, Codex, Cursor, or any other supported agent. The agent automatically installs the Glasser CLI and opens a browser window to sign in. After that, the agent handles the catalog itself, finds an endpoint for the job, checks how much the call will cost, runs it, and reports how much was spent. Most of these data sources are usually locked behind separate monthly plans reserved for SEO, sales, and company research tools. Glasser predicts that most teams would only need a few lookups from each source and would prefer to pay for those calls alone. There’s also an open-source skills library on GitHub facilitating jobs like competitor research, SEO audits, and prospect lists. The skills can be mounted on an existing agent setup, allowing a workflow that used to scrape pages to call a paid source. We tested it on two jobs, costing a total of $0.038 out of the free $1 credit. > An Apollo lookup on https://t.co/ZBzHukOHz8 returned a result in under half a second, with data on headcount, funding rounds, investors, a department breakdown, and 132 detected technologies, costing $0.026. > A DataForSEO search on our own domain found 2,995 keywords testingcatalog[.]com ranks for on Google in the US, and returned the top 100 with positions and search volume, for $0.012. New workspaces get $1 in credit to try it, and don’t require a card. Test it out 👀 https://t.co/qilS8V9YCu23d ago10K612113271.03%
You have been warned 💀 The internet isn't safe, and it never will be (and it never was) Cybersecurity risks aren't only determined by the level of cyber defense you have, but by the capabilities available for a cyber offense (Check NIST’s risk framework). It is not possible to make the system "100% secure", and that implies "continuous defense" will be required. In the future, 24/7 security agents will likely offer such defense. My concern is that many won't be able to access or afford such protection. Recent security vulnerability announcements are a direct consequence of the "Defenders Gap" (Cloudflare and Microsoft's reported malicious activity highlights the pressure on defenders). In fact, it's even worse now, as these incidents have shown. Organizations without access to advanced defensive tools may struggle to keep up. Cyber Defense is named "defense" for a reason, as these threats may pose a real monetary impact on the economy. If we think of programs like Daybreak as a way for AI labs to pay a "security tax," it becomes clear we have a huge supply-and-demand gap in this area. This is the gap between the protection governments need to provide to businesses and individuals and what they currently provide. Cybersecurity risks affect companies worth billions. Soon, tons of rogue agents will be running across the internet, trying to breach whatever they can (speculation). At some point, these incidents will be happening almost daily. This will affect normies' day-to-day lives, too. Yet, it is an unmatched opportunity to provide cyber defense solutions. YOU ARE NOT PREPARED!9h ago10K15911290292.28%
Mistral Large 4 jumps to the 6th spot on the Open-Weight Artificial Analysis Index. Le Chonk also scores 82 on the CyberGym-E2E (AA) benchmark. Similar performance appears across other Cyber benchmarks. Needs testing now 👀 Overall, it seems like a decent release. The biggest question is how quickly Mistral can ship new model upgrades to stay relevant.frozen16d ago10.0K120890121.49%
MICROSOFT 🔥: Miai silicon family is scaling to multi-gigawatt deployments in 2027 along with a new Maia 300 chip being announced! > Maia 300 is a complete intelligent system designed for the new types of AI workloads emerging today: Long-context reasoning, latency-sensitive conversational experiences and sustained, multi-step agentic tasks. > Maia 200 is now serving Microsoft production workloads, including Microsoft Copilot, and supporting the MAI team as they develop next-generation models, with GitHub Copilot and Microsoft Foundry to follow by end of year. A step closer towards full stack 👀frozen10d ago9.9K1121092121.46%
MISTRAL 🔥: Voxtral 3, a new family of speech-to-text and text-to-speech models, is now available! Voxtral Small Transcribe 3 landed in 3rd on the Artificial Analysis Speech-to-Text leaderboard, and Voxtral 3 TTS ranked in the top 12. Voxtral 3 comes with 👀 - Accurate transcription across 50+ languages. - Natural speech that adapts to the conversation. - Human-like intelligence at spoken command. - Pro Voice for distinctly accurate voice adaptation. Voxtral 3 models are now available through Mistral's EU-hosted API or for self-hosting.2d ago9.8K1207160261.72%
BREAKING 🔥: Meta unvealed Muse Charm, a new AI pin hardware device to allow users to communicate with their Muse agents from anywhere. Shipping will start in December - will we get Muse globally by then too? https://t.co/TZo25ZgGO2frozen29d ago9.8K3828290.60%
Nebius has expanded its AI Builder Program to Europe, offering $10M in combined compute grants from NVIDIA, Accel, and other European VCs. It's free to join and gives developers credits to run open models on Nebius; the European version also adds extra support for larger open-source projects that may need more compute or funding to grow. > Selected projects can be offered compute grants on Token Factory, Nebius's own service designed for running open models. > Accel and other European VCs provide founder funding to help teams turn an open-source project into a company. > Nebius and NVIDIA also run hackathons and demo days with communities such as AI Tinkerers, {Tech: Europe}, and ESRA.frozen3d ago9.8K723161101.05%
The team behind @readme has launched Restless to help companies make their APIs work better with AI agents. > Restless captures live API traffic from agents and turns their feedback into Linear issues or suggested documentation updates. > It enables companies to provide instructions on how agents should proceed after an API request fails. > Failed responses include fix steps and a direct link to the documentation, giving agents a clear next step. A tool-use RSI 👀frozen10d ago9.7K638143311.23%
fal released FLUX 3 Max 🔥 > FLUX 3 Max is a post-trained video model from fal Research and BFL. The team optimized inference to improve output quality. > It performs on par with Seedance 2.5, with video generation up to 22x faster. > The model scored 1,139 points on the Artificial Analysis image-to-video index. Max testing time! 👀3d ago9.6K944123161.34%
Drafted AI floor plan generator is now available in Autodesk Revit as a plugin! The plugin allows users to turn a house footprint and a room list into five floor plan options, built with real Revit walls, floors, roofs, doors, and windows. > It returns the results in about 45 seconds, according to Drafted tests. > Rooms can be removed and regenerated while the rest of the plan stays unchanged. > Teams can swap the generated doors, windows, and walls for their own saved components and edit every element by hand. https://t.co/c1kEn65kgzfrozen23d ago9.6K74462371.28%
Deel, the global payroll and HR company, has launched Akai after achieving $140M ARR, an AI agent platform that automates repetitive back-office work for finance, HR, and compliance teams. Akai can take a single screen recording with a spoken walkthrough and build a team of agents to run specified tasks on a schedule. > Akai captures what the user clicks and says, then builds that task into a workflow its agents can repeat on a schedule. > It is able to connect to legacy bank and government portals that have no public API. > Invoice statements, tax rates, and other exact values are handled by fixed formulas, with AI steps reserved for judgment calls. Akai is available in early access, and the company is offering $5000 in free credits!frozen24d ago9.6K580111220.96%
DigitalOcean has launched a public preview of Signals, Simulations, and Insights, and a private preview of Agent Data Services. With these tools, teams can now catch issues with the first affected user and test the fix on that exact conversation before releasing. > Signals can read live agent traffic and flag sessions that went wrong, like a tool call that succeeded but queried the wrong region. > Users can turn a flagged session into a test, and Simulations will replay the whole conversation with the proposed fix. > Agent Data Services allows the agent to review old support call recordings and documents to figure out what caused the problem.frozen10d ago9.5K615122191.04%
Pine AI launched Pine Computer, a cloud computer, harness, and runtime layer built for agentic tasks. On SaaS-Bench v1.1, Pine Computer has the highest checkpoint score in the publisher's table: 78.3%, against 74.3% for Opus 5 with Claude Code. It also completed fewer whole tasks: 27.4% against 31.1%. The benchmark covers 106 multi-step tasks inside 23 business apps, and the checkpoint score counts the steps passed along the way. Instead of simulating clicks and screenshots, Pine Computer can read web pages as structured data and report changes directly to the AI. Access is through a private beta waitlist from @PineAIAssistantfrozen13d ago9.5K6310161301.26%
Attio has introduced a new CRM architecture that lets AI agents work alongside revenue teams on the same customer data. Teams can ask their agents questions about the data they are working on and get an SQL-backed answer. > The SQL engine can run joins and calculations inside Attio, so one query can return a weighted forecast or quarterly win rate. > Attio launched connectors so that BI tools like Metabase or Looker can read live CRM data through a Postgres-compatible link. > Agent scale allows up to 100 million records per object, letting thousands of agents work through the same account at once.frozen7d ago9.4K75691341.33%
META 🔥: Muse Spark Open (Muse-Spark-Open-816B-A42B) is now available on HuggingFace! > "Muse Spark Open is a mixture-of-experts model with 42B active parameters, 816B total parameters, and a 1M-token context window." > "It’s built for agentic and coding workflows, from understanding large repositories to completing multi-step engineering tasks with tools." Benchmarks 👀frozen7d ago9.4K15011173182.12%
Krea 3 Ultra and Krea 3 Fast landed in the top 5 and 11th spots on the Artificial Analysis image-editing leaderboard! > Krea 3 Ultra placed 4th with a score of 1151, one point behind Google's Nano Banana 2.1. > The other four spots in the 5 five are all taken by OpenAI, Google, and MAI models. > Krea 3 Fast scored 1125, landing in 11th place, topping Nano Banana 2 by one point. At the same time, Krea 3 is also 10x faster than GPT Image 2 and 4x faster than Nano Banana Pro, with pricing starting at $0.03 per image at 1K and $0.06 at 2K.2d ago9.4K893112181.31%
ScaleDown has launched 3 small AI models for routine in-app jobs: Classify, Extract, and Summarize. All 3 models have a 1M-token input limit, support text or images, and are accessible through ScaleDown's API, as well as OpenRouter, Vercel AI Gateway, LiteLLM, and LangChain. > Classify - routes text by assigning user-provided labels, including a probability for each label. > Extract - can read files, collect the desired data, and output it as JSON fields with a high confidence score. > Summarize - can read long transcripts or reports and generate a short summary of up to 1,000 tokens.frozen9d ago9.2K765121291.34%
Black Forest Labs launched FLUX 3 Image Klein, its new open-weight image model! > "Customizable and fast. 5.6B parameters and runs on a single consumer GPU. Native 4K, bounding boxes, and pixel-perfect editing." > Developers can post-train and fine-tune FLUX 3 Image Klein. > It comes with a low-latency version for faster generation and self-hosting, as well as quantized versions - FP8, NVFP4, and INT8. > The distilled model with native 4K will also be available on BFL APIs. https://t.co/sxzrJG6nZUfrozen7d ago9.2K11111193241.83%
AIPOCH Open-Science released a new open-source desktop app that lets AI agents handle research tasks, from reading papers to analyzing data in Python and R. The company built the app to lower the barrier to research, letting anyone with curiosity and a research idea get involved. By saving the code, inputs, and steps that produced each artifact, it lets others rerun and analyze published papers. > Researchers can use models from OpenAI, Anthropic, DeepSeek, and others, as well as local models. > Projects can be exported as .science save files that can be opened on another computer and continued from. > By default, the app keeps files, projects, and research records locally on the researcher's computer. > The app can be installed on macOS, Windows, and Linux, and its source code is licensed under Apache 2.0. A .science file saves the whole session into a single transferable file. In one of AIPOCH's shared examples, the agents examined public data on 278 breast cancer patients and tested whether immune cell arrangement in a tumor helps predict survival. Anyone can open that file in the app, read the code behind each figure, and fork it into a new run. Browse the gallery 👀 https://t.co/nyn8nyBfCdfrozen6d ago9.1K831140561.69%
Sabi has raised $50M from Khosla Ventures, Accel, Initialized, DST, and Collab Fund, with Kevin Weil also participating. The company is building the Sabi Cap, a non-invasive wearable that aims to read brain signals and turn thoughts into text. > Sabi designs its own custom chips and neuroimaging sensors. > Its non-contact EEG chip, fabricated by TSMC, reads signals without touching the scalp. > A Brain Foundation Model is trained on neural data collected by Sabi itself. SMART HATS ARE HERE! 🔥frozen13d ago9.0K56181130.87%
Tuesday Labs introduced a Voxel Series-1 home robot, with the first batch of 999 units shipping in 2028. Voxel is being built with a special focus on how it behaves around humans while tidying a home, from returning things to where they belong to grabbing something from another room and bringing it back. > At launch, it will find misplaced items and care for houseplants. > Tuesday Labs has worked with Pixar advisors on its personality, acting like a butler-in-training who likes to keep things in order. > Founder Ros Shinkle previously wrote the algorithms that allow Boston Dynamics' Spot to pick up and place objects.2d ago9.0K615160181.11%
Voyager has launched a desktop app that lets AI agents work inside creative tools like After Effects, DaVinci Resolve, Blender, and Unity on a Mac. It reads project files, operates creative apps, and outputs editable results. > Video edits, motion graphics, and color grading in the apps creators already use. > 3D scenes in Blender and game prototypes in Unity. > Built-in skills, custom skills, and a memory that learns how each user works.frozen14d ago9.0K94861671.96%
Sesame is building support for Connectors, Skills, Scheduled tasks, and task management features! > Currently, the app is free; however, Sesame will likely introduce several paid plans soon. Pricing and feature set are yet unclear. > A hidden Activity tab holds sections for "Tasks", "Schedules," and "Skills". > Connectors currently include Gmail, Google Calendar, and Google Drive. > Thinking effort for the underlying model will be adjustable as well. I still use it as my default voice assistant 👀frozen22d ago9.0K843100141.24%
ContextQA has launched Ship, an AI-powered quality engineer that can autonomously find and fix bugs. When a bug report appears in Slack or Linear, Ship automatically reproduces the issue on the running app and records the evidence. It can then investigate the likely cause, using code, tickets, tests, and docs as context Findings can then be sent to an engineer or to Claude and Codex as a starting point for the fix.17d ago8.8K55380170.94%
Mistral ❤️ NVIDIA Mistral AI partners with NVIDIA to “develop a new generation of AI-native engineering workflows powered by NVIDIA accelerated computing.” Additionally, Mistral's models will be integrated with NVIDIA Omniverse libraries for GPU-accelerated 3D scene generation.1d ago8.8K1159292141.93%
AgentMail has launched AgentID, a sign-in option that allows AI agents to access apps with a verified email, similar to how 'Sign in with Google' works for people. Each agent signs into apps with its own verified email, linked to a human owner. Apps that previously blocked agent signups can now allow them and see the owner behind each agent. > 10 agents linked to one person appear as a single customer. > Apps can set limits for each owner and disable individual agents. > A human owner can be contacted if an issue occurs.frozen16d ago8.7K753171201.33%
Volantis has raised $88 million in Series A funding co-led by Lachy Groom and Abstract, with John Doerr, Susa Ventures, VXI Capital, and Triatomic Capital also participating. Volantis is building optical links inside chip packages, providing AI accelerators access to large pools of faster, lower-cost memory. > The company reports that memory bandwidth and capacity per chip can increase by several times compared to current designs. > Volantis aims for up to 10,000 tokens per second per user on models with more than 10 trillion parameters. > It expects coding agents to complete tasks in minutes or seconds, compared to hours on current hardware.frozen20d ago8.7K3714030.52%
Podonos launched Onepin, an AI-powered voice production agent that checks and fixes every voiceover line before it ships. The system tests every voice line for naturalness, clarity, and pronunciation against the team's standards. > Mispronounced names and terms are fixed directly at the word level. > Brand names and technical vocabulary are checked on a 4M-word pronunciation dictionary. > Each line is routed to the best fit out of 30+ tested voice models.frozen21d ago8.7K45140130.73%
Envato has launched “Burst Mode”, a feature that turns one idea or reference into up to 6 visual directions for 1 credit, up to 10x faster. Envato is a creative asset marketplace founded in 2006, offering commercially licensed photos, video, music, fonts, templates, and 3D to millions of creative professionals. > Users can start from a rough idea, a half-formed thought, or a reference image. > References, moodboards, and a creativity control from focused to wild steer each batch. > Users can vary promising directions with “More like this,” then polish and take them further.frozen15d ago8.5K39320200.75%
Inworld just launched Inworld Inference, an LLM inference API built for consumer AI apps, serving open models including Gemma 4, GLM, DeepSeek, and MiMo. > @inworld optimizes how each model runs to deliver faster responses for tutors, companions, games, and other consumer apps. > On Gemma 4 26B, median time to first token is 0.17 seconds, compared with 0.61 seconds for the median of the other providers on OpenRouter - 3.6× faster. > At the 95th percentile, time to first token is 0.31 seconds versus 2.42 seconds for the median of those providers - 7.8× faster. > Developers can also set requests to switch automatically to a fallback model if the first model fails or takes too long.frozen8d ago8.5K594121191.12%
NVIDIA 🔥: A new OpenShell Reference Architecture has been introduced along with a new NVIDIA DSX POD. AI factory deployment at a prepared site is now reduced from six months down to 14 days. This reduction saves tons of money for the industry. https://t.co/rnQt1Qp25D https://t.co/BIy9Zfhpa41d ago8.4K8711213121.60%
DigitalOcean is hosting the Open Intelligence Summit with NVIDIA in San Francisco today and streaming every session live on YouTube. The main theme of the event is that companies should be able to build their AI products on open models under their own control. > The keynote main speakers are DigitalOcean CEO Paddy Srinivasan and Adel El Hallak, VP of Product for NVIDIA AI. > OpenHands CEO Robert Brennan is also showing how agents can handle more of the software development cycle. > To close the day, Jeff Boudier of Hugging Face is hosting a panel about where open models could go next.frozen9d ago8.2K373100100.73%
Recursive has released Flint, its first open-weight coding model under the Apache 2.0 license, which is 2x faster than comparable alternatives. > Flint is a 40B parameter mixture-of-experts model for local coding. > Recursive reports achieving more than 220 tokens per second on a 64GB RTX Spark. > Recursive's Eureka Machine, a system for running AI research experiments, wrote custom GPU code that sped Flint up by 55% on RTX Spark without changing the weights.1d ago8.1K68661161.20%
Arceus has launched a law firm that uses AI agents and licensed attorneys to handle commercial contract work for startups and scale-ups. The firm can review, redline, draft, and negotiate MSAs, DPAs, NDAs, order forms, and vendor agreements end-to-end. > Clients can send contracts and legal requests to Arceus directly from Slack. > Every job gets a flat price that the client approves before work begins. > AI agents gather the context, and a licensed attorney reviews every output.frozen21d ago8.1K38261110.72%
Zenzap raised $44 million in a recent round led by investors including Bessemer, Lightspeed, and The Garage, and launched AI agents in its workplace chat app. Zenzap is a work chat app for teams that typically coordinate on their phones, such as restaurant and store staff. > Inside the app, groups include both humans and AI agents that collaborate on tasks. > Agents can connect to more than 400 apps, including CRMs and shift-scheduling tools. > Users can schedule a recurring task by describing it in a chat message, and any changes to data on a connected service require the user’s approval.frozen8d ago7.9K38260180.81%
Bland has launched a public demo of Blandie, its new tool that can create branded voice and SMS agents for businesses based on their website. Users get a verified chat with that business’s name and logo, connected to an agent that can handle bookings and payments. > Messages are sent by a verified sender with the company’s name and logo, as RCS on Android and iMessage on iPhone. > Answers are based on the content from the provided website, where the agent is also capable of responding with website screenshots and images. > The agent can remember if a customer has already called the business’s @usebland phone number and refer to it in the text chat.frozen7d ago7.9K48460221.01%
Coco Labs launched Vyb, its first design model, which can build layered, editable marketing graphics from a brand's website. Vyb places each element on its own layer, so you can change a headline without regenerating the whole design. > Users can send Vyb their company's website and have it pick up the brand's logo and exact hex colors. > Product photos can be added into the design at full resolution without being regenerated. > Produced designs can be exported to Figma while retaining every layer intact.2d ago7.7K33390110.72%
GMI Cloud has raised $668M in a Series B round led by ARCHIV, with participation from NVIDIA and a credit facility led by CTBC. @gmi_cloud provides GPU compute and inference across the U.S. and Asia-Pacific on a single platform, supporting first-time builders and advanced AI labs. The platform is built for training on the latest GPU clusters, large-scale inference, and deploying agents directly to production. More compute incoming! 🔥 https://t.co/9FwxgD9POqfrozen22d ago7.7K2906170.56%
General Compute has signed a multi-year partnership with @positron_ai to offer Positron's Atlas inference systems to its cloud customers. General Compute buys and runs specialized AI chips alongside GPUs, allowing AI companies to rent processing power and use Positron hardware without buying it. > GPUs handle the input process for each request, so Positron chips generate the output more efficiently. > Positron’s chips are built with a focus on fast memory access since it is one of the main bottlenecks for token generation. > Atlas is already available inside Oracle Cloud Infrastructure, with General Compute also committing to integrate Positron's next-generation Titan systems.2d ago7.6K4738190.89%
Ampersand has raised $15M in a Series A led by Bessemer Venture Partners, with Lauri Moore joining the board, and has made its platform generally available. Ampersand lets AI agents read and update records inside Salesforce, SAP, NetSuite, and Workday, even when every customer has customized it differently. > Read actions can pull data on a schedule or on demand, with backfills. > Write actions are able to create and update records in the customer's system. > Subscribe actions deliver changes in real time as records update.frozen16d ago7.6K30110090.66%