← Back to leaderboard
zarairahh

Zara

Verifiedactive
@zarairahh

AI Explorer & Creator | DM for collaborations 📧 contactzarairahh@gmail.com CPP • @itsPolloAI, @higgsfield, @capcutapp, @openart_ai, @kling_ai

RosterAI Content Creation & Video Gen
14,728 followers · 980 following · Last polled 20m ago
Perf. Score
73.1
Rank #54
Value Score
97.2
#5 of 268 priced
Median views
17K
p25 14K · trailing 7d
Followers
15K
+0.3% 7d
ER (impr.)
1.74%
eng ÷ impressions
ER (foll.)
1.95%
eng ÷ followers
Est. CPM
$1.21
QT-rate basis
WoW views
-35.7%
▼ falling

Rates & value · est. from 17K median organic views

underpriced
Quote tweetbasis
$20
$1.21 / 1K views
Post
—
no CPM
Thread
—
no CPM
Retweet
$17
no CPM
Value Score 97.2 · rank #5 of 268 priced$71.17 / 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 (17K), not the mean (16K) — so a single viral post doesn't inflate a creator's reach. The p25 floor (14K) is the level 75% of posts clear.

Recent posts · ranked by views

PostPostedViewsLikesRepostsRepliesQuotesBookmarksER
Somewhere Between Sleep and Seoul. Made with Seedance 2.5 on @capcutapp #capcutcpp https://t.co/caWoLis48Efrozen8d ago108K669966021690.92%
I think a video can look finished without looking perfectly clean. In this piece I created with Pexo, the torn edges, tape, and crumpled silver texture give the images a physical quality. The red takes up very little of the frame, but it’s hard to miss. That’s the part I like: the materials have as much presence as the images themselves. A smoother version would have a different mood. It’s also why visual references matter when developing a video with Pexo. Alongside the story and detailed direction, they help communicate what “finished” should look like for that particular piece. Here, the roughness belongs in the result. https://t.co/PgSQDDjDrhfrozen13d ago49K11227
33
0
4
0.36%
What if your everyday foundation could look this effortless? ✨ Cream-balm texture. Natural finish. Skin that still looks like skin. Made with @alibaba_cloud Wan 3.0 on @openart_ai #OpenArtAdAwards https://t.co/DNttwcvQAAfrozen16d ago31K27130530181.19%
This AI cinematic recreation of 1950s London is absolutely insane 🔥 A crowded market. A mysterious key. A hidden suitcase. One secret waiting to be uncovered. AI is getting seriously unreal with Seedance 2.5 on @openart_ai 🤯 https://t.co/5iY2xyLIMn3d ago25K24267570431.61%
She had a tiny umbrella, but a big heart. ☔💛 One small act of kindness. One unexpected ripple. A reminder that we never know how far our actions can reach. Made with Seedance 2.5 on @newtake_kr #newtake #nextscenechallenge https://t.co/z9ziFmHwjxfrozen28d ago25K18933640221.22%
Paradise looks better when you wear a little confidence. 🌊✨ Created with Seedance 2.5 on @openart_ai Prompt: Create a photorealistic 30-second cinematic tropical travel vlog montage set in Hawaii, featuring the SAME beautiful 20-year-old adult East Asian-American woman throughout the entire video. Use the uploaded reference image as the primary visual reference for her overall appearance, hairstyle, styling, outfit, and proportions. Preserve strong character consistency from beginning to end: same face, dark hair, natural skin texture, facial proportions, body proportions, hairstyle, and recognizable appearance. Do not change her identity or appearance between scenes. CHARACTER & STYLE: A beautiful 20-year-old adult woman with a youthful but clearly adult appearance, dark black-brown hair, natural makeup, soft expressive features, relaxed confident personality, and an effortless American vacation aesthetic. She has a fashionable, feminine, slightly bold tropical-resort style. Her expressions should feel candid and natural rather than posed. WARDROBE — KEEP CONSISTENT: She wears a cream/beige open-knit crochet off-the-shoulder beach top, loosely draped over a simple white fitted camisole/bikini-style top, paired with high-waisted white distressed denim shorts. The crochet top is airy, oversized, and slightly sheer as naturally created by open-knit fabric, with tasteful resort styling. She wears minimal delicate accessories. HAIRSTYLE: Dark hair gathered into a low, slightly messy bun, with loose strands naturally framing her face. Two small tropical flowers, one white and one pink, are tucked into the bun. Keep the hairstyle and flower placement consistent throughout the daytime scenes. Natural wind movement in her hair. IMPORTANT CONSISTENCY: The woman must remain the exact same person in every shot. Maintain identical facial features, skin tone, hairstyle, flower placement, body proportions, and overall appearance. Keep the same cream crochet top, white underlayer, and white denim shorts throughout the daytime scenes. Do not randomly change her clothing, hairstyle, age, face, or body shape. Natural realistic anatomy and realistic hands. VIDEO FORMAT: 4K photorealistic cinematic video, 24fps, 30 seconds, 35mm film aesthetic, subtle authentic film grain, realistic handheld travel-vlog cinematography, natural autofocus adjustments, shallow depth of field, realistic motion blur, soft highlights, natural skin texture, warm vintage color grading, atmospheric lighting, cinematic composition, documentary-style realism. OVERALL VISUAL MOOD: Dreamy luxury Hawaii summer vacation, warm tropical atmosphere, authentic travel diary, feminine and fashionable, relaxed confidence, natural candid moments, beautiful ocean scenery, golden sunlight, nostalgic 35mm film look. Avoid looking like a commercial with overly perfect poses; make it feel like an expensive but authentic personal travel vlog. SCENE 1 — HAWAII ARRIVAL & TROPICAL CITY WALK — 0–4 SECONDS: Beautiful Hawaiian morning. The woman walks casually through a bright tropical street wearing the exact reference outfit: cream crochet off-the-shoulder top, white fitted underlayer, high-waisted white distressed denim shorts, and tropical flowers in her low bun. Palm trees, tropical plants, colorful architecture, small cafes, and casually walking people in the background. Camera begins behind her at waist level, following naturally as she walks, then smoothly moves around to a three-quarter front angle. She turns slightly toward the camera and gives a genuine relaxed smile. Warm sunlight catches the edges of her hair. Natural breeze moves the crochet fabric and loose hair strands. SCENE 2 — BEACH DISCOVERY — 4–8 SECONDS: She arrives at a wide beautiful Hawaiian beach. She walks barefoot along the shoreline in the same outfit, carrying her sandals casually in one hand. Crystal-blue ocean waves gently reach her feet. Capture cinematic low-angle shots of her bare feet stepping through wet sand, water reflections, and small waves. Cut to a medium-wide shot of her walking toward the ocean with distant volcanic mountains behind her. She looks happy and relaxed, enjoying the moment. Realistic ocean movement, natural sunlight, subtle lens flare. SCENE 3 — PALM TREES & OCEAN CLIFF — 8–12 SECONDS: Start with a cinematic worm's-eye view looking upward through tall tropical palm trees. Golden sunlight streams through the leaves with subtle natural lens flares. Smooth transition to the woman standing near a safe rocky ocean viewpoint, still wearing the exact same reference outfit and flower-decorated bun. Wind gently moves her hair and crochet top. She looks toward the Pacific Ocean with a peaceful expression. Camera slowly pushes in from behind her shoulder into a cinematic side profile. SCENE 4 — BEACHFRONT CAFE — 12–16 SECONDS: She sits at a cozy stylish beachfront cafe beside a large open window. Same outfit and hairstyle. A fresh tropical drink sits on the table beside her sunglasses and small personal accessories. She casually takes a sip while looking outside at the ocean. Capture intimate cinematic close-ups of her hand holding the drink, condensation on the glass, tropical fruit, sunlight across the table, ocean waves outside, and her relaxed thoughtful expression. Warm sunlight creates soft highlights on her face. Authentic candid travel-diary feeling. SCENE 5 — OCEAN ADVENTURE — 16–20 SECONDS: She is now enjoying calm turquoise Hawaiian water on a small surfboard, wearing the same tasteful white underlayer with the cream crochet top removed for practical water activity while keeping the same overall styling and character appearance. Keep the white high-waisted shorts only if physically realistic for the water activity; otherwise use tasteful coordinated white resort swimwear. Do NOT alter her face, hairstyle, or body proportions. Camera moves at water level around her as gentle waves pass by. Sunlight sparkles across the turquoise ocean. Tropical coastline and mountains appear in the background. She laughs naturally toward the camera, creating an authentic vacation moment. SCENE 6 — HAWAIIAN NIGHT MARKET — 20–24 SECONDS: Transition into a vibrant Hawaiian night market filled with warm hanging lights, tropical decorations, food stalls, lanterns, colorful signs, and natural crowds. For this evening scene, she changes into a simple elegant white resort dress while maintaining the exact same face, hairstyle, and recognizable appearance. She walks through the market, samples tropical fruit and local street food, and reacts with a genuine delighted smile. Cinematic close-ups of tropical fruit, food preparation, glowing lanterns, her expression, and colorful nighttime bokeh. Realistic handheld camera movement. SCENE 7 — GOLDEN SUNSET — 24–27 SECONDS: Wide cinematic shot of the SAME woman standing barefoot at the ocean shoreline during golden sunset. She wears a flowing lightweight white resort dress. Orange, pink, and golden colors fill the sky and reflect across the water. Gentle waves move around her feet. She stands peacefully looking toward the horizon. Camera slowly pulls backward, creating an emotional cinematic travel-film ending. Natural wind moves the dress and her hair. SCENE 8 — HOTEL NIGHT REFLECTION — 27–30 SECONDS: Nighttime luxury high-rise hotel room and balcony overlooking sparkling Hawaiian city lights. The same woman wears an elegant simple white resort outfit. Warm interior lighting contrasts with the cool tropical night outside. She stands on the balcony while sheer curtains move gently in the warm breeze. Cut to a final intimate close-up of her relaxing on the hotel bed, looking warmly into the camera with a peaceful, confident smile. End with a subtle cinematic focus pull from her eyes to the sparkling city lights beyond the balcony. CAMERA: Authentic handheld travel-vlog cinematography, subtle natural camera shake, smooth gimbal-like movement when appropriate, slow push-ins, gentle tracking shots, occasional POV shots, realistic autofocus breathing, natural rack focus, shallow depth of field, cinematic close-ups, wide establishing shots, low-angle beach shots, water-level camera movement, subtle motion blur, realistic lens characteristics. LIGHTING: Natural Hawaiian sunlight, soft morning light, warm golden-hour sunlight, realistic tropical reflections, subtle lens flares, atmospheric evening lighting, warm cafe lighting, realistic nighttime bokeh. No artificial-looking studio lighting. COLOR & TEXTURE: Warm nostalgic 35mm film color grading, soft tropical highlights, realistic skin tones, subtle film grain, natural contrast, gentle vintage character, cinematic depth of field, realistic fabric texture, realistic ocean reflections, detailed tropical vegetation. REALISM: Photorealistic live-action appearance. Realistic human anatomy, realistic hands and fingers, realistic hair strands, realistic fabric physics, realistic water physics, natural walking and body movement, believable facial expressions, natural blinking and breathing. Make the footage feel like it was captured by a professional travel filmmaker on a real Hawaiian vacation. NEGATIVE PROMPT: No cartoon, no animation, no CGI appearance, no plastic skin, no artificial beauty filter, no overly airbrushed skin, no distorted face, no changing identity, no inconsistent facial features, no changing ethnicity, no changing age, no body distortion, no exaggerated body proportions, no unnatural anatomy, no extra fingers, no missing fingers, no malformed hands, no duplicate person, no duplicate limbs, no warped clothing, no floating objects, no unnatural hair, no hairstyle changes without the specified outfit transition, no flickering face, no facial morphing, no blurry face, no unnatural expressions, no stiff movements, no robotic motion, no oversaturated colors, no excessive lens flare, no unrealistic lighting, no fake-looking ocean, no text, no watermark, no logo. FINAL PRIORITY: Character consistency and photorealism are the highest priorities. The uploaded reference image should strongly guide the woman's appearance and daytime outfit. Preserve her recognizable appearance across every scene while creating a seamless, dreamy, authentic Hawaii luxury travel vlog.frozen1mo ago24K19213502581.32%
Coding agents can write a lot of code now. But someone still has to check if it actually works, reproduce the weird bugs, and figure out what went wrong. Ship is a pretty natural companion to coding agents, handling the testing and debugging work that comes after.3d ago21K1817524011.35%
Created with Kling 3.0 on @Kling_ai Prompt: Create a 15-second high-fashion editorial film starring the adult character in <<<image_1>>> . Preserve their exact appearance, outfit, accessories and visual style. Render the entire film in the reference’s aesthetic. Use a minimal studio with a reflective floor, bold directional lighting and rhythmic photographic flashes. Show exactly 10 distinct fashion poses, each with a different silhouette and camera angle: 1. A confident standing pose, filmed from floor level with dramatic wide-angle perspective. 2. One hand beside the face, captured in an extreme side-profile close-up. 3. An asymmetric seated pose, seen directly overhead as the camera rotates. 4. One leg extended toward the lens, the foot or reference footwear dominating the foreground through foreshortening. 5. A quick, fluid leg cross, captured from a low diagonal angle. 6. Fingertips catching the edge of an existing sleeve or collar, framed in a tight detail shot. 7. A rear three-quarter turn with a deliberate glance back into the camera. 8. A forward lean toward the lens, ending with the palm briefly covering it. 9. An elongated upward stretch, revealed in a sharply tilted wide shot with strong side lighting. 10. A commanding full-body hero pose, looking down into the lens as the camera rapidly pulls back. Keep the performance fast and rhythmically tight. Flow naturally between poses, holding each for only a fraction of a second while keeping its silhouette clearly readable. No repeated poses or additional pose changes. Connect angles through whip pans, movement-matched cuts, snap zooms, flashes and lens occlusions. Orbit against the character’s turn, sweep from their floor reflection toward their face and plunge from overhead to floor level. Alternate intimate details with dramatic wide compositions. Accelerate between poses and brake sharply on each hold. Keep the sensual gestures quick, confident and fully clothed. Hair and fabric react naturally. No slow motion, lingering shots or gradual camera drift. Build momentum toward the tenth pose, then stop sharply for a crisp final hold. Stable identity and anatomy throughout. No outfit changes, added accessories or text. Sound: a driving 140 BPM fashion beat, rhythmic shutter clicks and subtle fabric movement.frozen29d ago20K239385811592.53%
This doesn’t even look AI-generated 🔥 Higgsfield Genjutsu https://t.co/LVudBI5j6Kfrozen4d ago19K22239600471.99%
AI is getting scary realistic. 🔥 https://t.co/RruvGg9J9Mfrozen6d ago17K19857420441.97%
This is probably one of the clearest examples yet of where AI in healthcare is headed. Not just answering questions, but actually taking care of the whole visit. And starting with acne honestly makes a lot of sense.3d ago17K1454932001.36%
48% thought Griffin was a real human. That’s a pretty wild milestone for AI. Griffin from Tavus can see, listen, speak, and react in real time. See what they built → https://t.co/5itvf0LznW Congrats @tavus and @hassaanraza https://t.co/2IiAotuV98frozen6d ago15K1273034081.30%
The CRM problem was never a shortage of AI Agents keep getting handed a customer record built for humans filling in fields by hand Lightfield starts from the other end → build the record for agents first, then let agents work on top $47M Series A led by a16z Been working with the Lightfield team, opinions are minefrozen28d ago15K916280130.90%
Nano Banana 2.1 is LIVE on @openart_ai Faster generations. Better quality. More control over complex prompts. Image Prompt: Create a photorealistic portrait of her in a refined side-profile-inspired three-quarter angle, with her head naturally turned toward the camera and making direct eye contact. Clearly defined SIDE PART — NEVER a center or middle part. She has extremely long, rich Dark Mocha Mousse curly hair, dramatically moved by a strong natural breeze, with realistic flyaways and loose strands. No ribbons, bows, headwear, hats, or hair accessories. Soft-glam Korean-inspired makeup: visible natural pores, luminous realistic skin texture, rosy blush, subtle brown eyeliner, softly defined eyes, rosy gradient lips, and nude espresso nail art. She wears a magenta-soft-purple thin-strap mini dress with a subtle heart-shaped neckline barely visible at the chest. Gold fairy-wing pendant necklace with matching earrings. ONE HAND is naturally raised in front of her face, gently pushing or clearing a hibiscus flower away from her line of sight. Fingers and joints must be anatomically realistic. Her other hand is completely outside the frame. Expression: calm, intimate, slightly curious, natural and emotionally subtle. EYE-LEVEL CAMERA positioned directly among dense blooming soft-pink and white hibiscus flowers, photographing THROUGH the flowers. Foreground petals, leaves, and stems are very close to the lens and naturally overlap the outer edges of the frame, creating an intimate organic floral frame. Flowers must not obscure her eyes or facial features. NO WIDE VIEW. The subject fills approximately 85–95% of the frame. ONLY HEAD TO CHEST VISIBLE. VISUAL STYLE: Warm nostalgic historical-memory aesthetic. Authentic vintage Canon analog-film character, subtle natural film grain, delicate halation, slightly faded tones, gentle light leaks, realistic exposure imperfections, soft natural shadows, authentic raw human skin texture, and warm soft daylight. Strong wind visibly moves the curls and loose strands. Mild natural motion blur ONLY on moving hair and nearby flower petals. The final image should feel like a candid, emotionally intimate photograph captured on vintage film rather than a polished studio portrait. NEGATIVE PROMPT: center part, middle part, ribbons, bows, headwear, hat, full body, waist-down, staircase focus, wide shot, distant subject, excessive background, artificial bokeh, studio lighting, smartphone photography, modern digital DSLR look, CGI, cartoon, anime, illustration, plastic skin, beauty filter, over-retouched skin, excessive facial smoothing, face swap, distorted facial anatomy, unnatural hands, malformed fingers, extra fingers, missing fingers, floating limbs, flower covering face, obscured eyes, closed eyes, facial distortion, unnatural pose, watermark, text, logo.22h ago14K25611380292.44%
The first generation is not always the version you want. That's why I like this workflow. GPT-6 Astra gives me a 3D white model to start from. CapCut PC + Seedance 2.5 turns it into commercial footage. Then I can actually work with the footage. Need a different pace? Edit it. Need a stronger frozen moment? Adjust it. Need more footage? Use AI Extend. Need another editing pass? Edit Pilot. The workflow leaves room for iteration instead of forcing a one-shot result. #GPT6Astra #CapCutPC #AIVideo #Seedance25frozen26d ago13K12718620131.63%
OpenScience is making scientific AI feel much more open. Model-agnostic, BYO ChatGPT/Codex, 100s of research skills, 50+ databases, and fully open source. If you’re building or researching with AI, worth checking out. GitHub: https://t.co/H6aVk4poZOfrozen11d ago13K13025260101.45%
40–400x cheaper is more than a cost reduction. It changes the set of problems that make sense to solve with AI. When intelligence costs almost nothing, you don't just build cheaper AI products. You build things that weren't economical to build before.frozen22d ago13K882327011.10%
AI agents are getting smarter. Now we need to understand how they behave together. Agora-2 gives multiple humans and agents the same world and lets their actions shape what happens next. https://t.co/xRkDXcbV5E13d ago13K1202936071.54%
The next frontier of AI isn’t just intelligence in the cloud. It’s intelligence that understands physics, cause & effect, and the physical world. Odyssey-3 is a glimpse of what happens when one world model can control robots, humanoids, cars, drones and even generate worlds for AI to learn in. This is big.frozen22d ago12K92336021.07%
Four open models. One picker. Builders chose. GLM 5.3 Flash took the biggest share of Bolt Forge prompts. Up to 50x usage on Flash. No benchmark needed when people can just build. https://t.co/DL9JeE4IkAfrozen20d ago12K10025270151.34%
Higgsfield casually dropping this like it’s nothing lol.frozen21d ago12K10123210141.32%
A director doesn’t just decide what the audience sees. They decide what the audience doesn’t see. That’s one reason spatial planning feels important for AI video. GPT-6 Astra lets me map the environment and establish camera + character movement within the Viewport. I can then manually bring the 3D scene into CapCut PC, use Seedance 2.5 to generate the footage, and use Edit Pilot to keep refining the final sequence. AI can help create almost unlimited possibilities. The creative challenge may become knowing which possibilities to leave out. #GPT6Astra #AIVideo #Seedance25 #CapCutPCfrozen29d ago12K10621510121.58%
Spent some time trying A1 and the first thing I noticed was how much more natural the interaction feels. It’s not just a talking face. The character can move around, interact with objects, and you can interrupt or change your request while it’s still responding. Pretty wild to see this happening in real time. @VivixLabs_HQfrozen7d ago12K1333237051.73%
If you already use ChatGPT, Claude, or Cursor, this is worth checking out. Pollo MCP connects Pollo AI directly to your workflow for image and video generation. New users can claim 40 credits to experiment. Try Pollo MCP: https://t.co/5bQKWjPEuxfrozen18d ago12K1483038091.92%
AI video has been getting better, but the workflow is still way too complicated. @videoclaw takes a different approach: just tell it what you want and let the agent handle the editing, generation, B-roll, captions, voice, etc. Basically, vibe coding for video. https://t.co/CYGB09xn90frozen21d ago11K73232070.99%
The smallest CX hiccup can become a big problem when it hits at the worst possible moment. Cresta helps customer service teams respond faster, solve issues efficiently, and keep the experience on track when it matters most. Because even Beast Mode has no patience for bad CX.frozen14d ago11K1012723011.33%
Create an AI influencer once, then bring them into any trend. Simple concept, big creative potential.frozen5d ago11K1233533081.80%
AI infra is getting hilarious. You can spend millions trying to build the video stack yourself… or just plug into Higgsfield and start shipping 😂frozen15d ago11K1282543021.80%
AI can create the content. Your brand needs to stay consistent. Templafy connects your AI workflow to the templates, assets, and brand rules your team already uses, turning ideas into polished, ready-to-share decks.frozen21d ago11K972035011.42%
GPT Image 2.5 just landed on Pollo and it's fast. Genuinely didn't expect this level of speed jump from a new model. Also it's unlimited for 10 days right now, so I've barely stopped generating since I found out.frozen29d ago11K931440021.39%
This is probably the right way to use AI videos. @itsPolloAI https://t.co/PtlOGwS8TIfrozen27d ago10K971642041.52%
we've spent a lot of time making agents smarter. the next bottleneck is making them easier to actually run. DO managed agents brings the runtime, tools, sessions and inference together so you don't need a mini infrastructure project for every agent you want to deploy. https://t.co/qKmNjrY0skfrozen15d ago10K911325061.29%
OpenAI x Higgsfield making creative work continue while you’re away is seriously next level.frozen8d ago10K1053139021.76%
building with AI gets a lot more fun when you don't have to worry about running out of usage. @boltdotnew just launched Forge with up to 50x more usage. you can try an idea, change it, scrap it, and start over without thinking twice. just build until you get something you actually like.frozen23d ago9.9K1021941011.65%
Okay, I seriously didn’t think this would be THIS fun 😭 Gave @noiz_ai two simple photos, and somehow they turned into an entire little story. It’s wild how a basic idea can suddenly feel so real once you see it come alive. #noizai #trend https://t.co/In60z0uWt9frozen5d ago9.6K14728410152.40%
2B parameters doing what we used to expect from much larger models. Local coding agents are evolving fast.frozen26d ago9.6K871033011.36%
This could easily be a product launch video. One model rotates through several colorways while the scene changes with it — outfit, background product, color swatch, everything. GPT Image 2.5 builds the visual variations. Seedance 2.5 turns them into motion. CapCut PC turns the generated footage into a polished final edit. And soon, GPT Image 2.5 will be available directly in CapCut through Design Studio & AI Image. #GPTImage25 #CapCutPC #AIVideo #Seedance25frozen22d ago9.4K11529510182.26%
This AI fashion transformation is absolutely insane 🔥 4 luxury looks. One seamless transformation. Pure cinematic magic. Created with Seedance 2.5 on @openart_ai https://t.co/992CVgFYjkfrozen10d ago9.3K13626400222.41%
Support gets a feature request every week. Instead of sending it to engineering, they can use Base Code to turn it into a real PR. Engineering reviews it. Customers get what they asked for. It’s kind of insane that this is becoming a normal workflow. Try Base Code: https://t.co/9iwgHB8Zo7frozen10d ago9.0K10441280102.04%
I still can’t believe Higgsfield Genjutsu matched the original this closely.frozen16d ago8.8K972226021.67%
50+ models in one API, no subscription lock-in, and up to 50% off. Hard to ignore this for builders.frozen21d ago8.8K791319001.27%
This Credits Reset is such a nice surprise for everyone creating on Higgsfield.frozen9d ago8.5K14040310242.78%
Excited to share that I was selected as an Official Tastemaker for OpenArt Arena. I got to take part in blind-testing some of the leading creative AI models, evaluating their outputs and helping contribute to the final rankings. It was a really interesting experience to see how different models performed without knowing which one I was testing. The first rankings are now live. Check out which models came out on top: https://t.co/7XJkifKWEKfrozen19d ago8.3K13229440202.70%
2B parameters on a phone, handling reminders, texts, searches, and calendar tasks. We’re getting surprisingly close to truly local AI assistants.frozen25d ago7.9K80328001.40%