← Back to leaderboard
sheldon056

Duet | AI

Verifiedactive
@sheldon056

Exploring Tech & AI | Sharing ideas, trends & witty randomness | DM for collab and paid promotions. ✉️ shmurphy056@gmail.com CPP: @higgsfield

RosterAI Content Creation & Video Gen
14,057 followers · 857 following · Last polled 3h ago
Perf. Score
55
Rank #184
Value Score
80.8
#53 of 273 priced
Median views
5.9K
p25 4.5K · trailing 7d
Followers
14K
+0.7% 7d
ER (impr.)
1.24%
eng ÷ impressions
ER (foll.)
0.86%
eng ÷ followers
Est. CPM
$5.09
QT-rate basis
WoW views
+93.7%
▲ rising

Rates & value · est. from 5.9K median organic views

underpriced
Quote tweetbasis
$30
$5.09 / 1K views
Post
—
no CPM
Thread
$13
$2.21 / 1K views
Retweet
—
no CPM
Value Score 80.8 · rank #53 of 273 priced$283.02 / 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 (5.9K), not the mean (9.7K) — so a single viral post doesn't inflate a creator's reach. The p25 floor (4.5K) is the level 75% of posts clear.

Recent posts · ranked by views

PostPostedViewsLikesRepostsRepliesQuotesBookmarksER
A few lines of creative direction can cover a lot of editing work. That’s the appeal of CapCut Web × Codex. Describe the captions, filter changes, and transitions you need, and let Codex call CapCut’s editing capabilities and libraries for music, sound effects, and visual effects. ChatGPT helps understand what you want the video to communicate. CapCut brings the tools and resources to the task. It’s a more approachable way to handle the finishing details and reduce the time they take. #capcutweb #capcutchatgptplugin #capcutcodex #capcutmarketing2d ago52K2307190330.56%
A little cafeteria chaos turned into a whole movie. The switch from messing with him to protecting him… And that head pat at the end? Absolutely unfair. Made this with @Flovaai. Try the Skill ↓ https://t.co/ZsF8JzE1QJfrozen24d ago16K1823230
4
1.29%
What if Walter White and Gus Fring were monkeys? I used Pollo Recast on @itsPolloAI to transform these iconic Breaking Bad characters into monkey versions while keeping their signature looks and personalities. The result is wild. https://t.co/Jvm5Tmrcas18h ago16K1804732061.67%
“Dime Dime Baby” came on, and she understood the assignment 💅💃 Not a dance session, just a full-blown solo concert 😂🔥 The crowd got a show they never signed up for! 😭 Made with @noiz_ai #NoizAI https://t.co/6b1edupLy13d ago15K1302200121.06%
What happens when football’s greatest rivalry moves from the pitch to the rap booth? Messi and Ronaldo under one mic, turning years of legendary rivalry into an unexpected rap performance with FlovaAI. Made with @Flovaai. https://t.co/nU7d65uwZ8frozen18d ago14K137218051.13%
Pollo MCP makes AI video workflows feel way more connected. I plugged Pollo directly into ChatGPT through MCP, so I could stay in ChatGPT, direct the creative process, and generate the visuals through Pollo without switching between tools. ChatGPT → story + direction 
 ChatGPT Image 2 → storyboard + characters 
 Pollo MCP → image/video generation
 Seedance 2.5 → brings it to life. I used this flow to create a cinematic 15-second chips ad with 3 consistent characters. Story → storyboard → video, all in one flow.frozen29d ago12K10726190171.38%
The Flova × Seedance 2.5 sale is here for 7 extra days! Get up to 80% OFF and secure your discounted Seedance 2.5 rates for 365 days. • 480p: From $0.022/sec • 720p: From $0.044/sec More deals: • Image 2.5: From $0.011/image • Wan 3.0 Prime: From $0.022/sec Offer ends October 27. More creativity, lower costs. Make the most of this final extension!1d ago11K110217021.16%
Batman just pulled up in GTA mission mode. 🦇 Made with FlovaAI’s GTA Skill, “Third-Person Gameplay Director.” It genuinely feels like real gameplay. Want to make your own? Skill link in the comments. https://t.co/UKYoLJ4Upifrozen29d ago9.6K1001170221.46%
#PixVerseWorldModel What if AI-generated video wasn’t something you simply watched, but something you could step into? That’s what makes @PixVerse World Model interesting. Instead of generating a fixed sequence and watching it play out, the WASD experience lets you move through the generated environment and experience the scene from different perspectives. You can explore the world, interact with characters and objects, and use prompts to influence what happens around you. The experience becomes more dynamic because your actions are part of the process. Move somewhere new, try a different prompt, and see how the generated world responds. It’s a different way of thinking about AI video: not just creating a finished clip, but creating an environment that you can continue to explore. As AI-generated worlds become more interactive, video creation could start to feel less like watching a movie and more like entering a world of your own. Try it: https://t.co/HjWWO0uuSnfrozen1mo ago9.2K106122021.42%
VFX artists are going to hate this. Made with Higgsfield + Seedance 2.5. https://t.co/62YNnjKmGvfrozen26d ago9.2K1236290382.14%
POV: Two photos. One dreamy little universe. Turned a couple of simple snaps into a full cinematic AI video with @noiz_ai. Drop yourself, your bestie, or your favorite character into a dreamy pink APT-inspired world and watch it come alive. Playful, magical, and straight out of a music video — all from just two photos. #noizaifrozen22d ago8.7K88118031.26%
AI is getting scary good. Is this creativity or are filmmakers cooked? 💀 Made with Seedance 2.5 on Higgsfield. https://t.co/KfqVyQVxXZfrozen7d ago8.0K1175270172.08%
Imagine spending months making what AI does in minutes. Made with Seedance 2.5 on Higgsfield. https://t.co/REuRedX346frozen22d ago7.7K67310051.11%
Recreated one of Thomas Shelby’s most iconic scenes with AI. Higgsfield Genjutsu is crazy. https://t.co/4MYG3jGXnsfrozen16d ago7.3K1243240112.21%
Your character. Any motion. Endless possibilities. 🐱⚔️ Flova’s MORPHJUTSU Skill lets you bring any character to life using a reference image, motion video, and audio. Just add your inputs and watch Flova transform the scene. Made with @Flovaai https://t.co/zxxScTZ9lN2d ago6.9K632200211.54%
Love survived. Everything else didn’t. Generated with Seedance 2.5 on @frameo_ai #MadeWithFrameo Prompt: Create a photorealistic, emotionally devastating cinematic video, approximately 24 seconds, 16:9, 24fps. A post-apocalyptic love story told entirely through visuals, without dialogue. Two infected lovers discover each other inside an abandoned train station. Their bodies have succumbed to the infection, but some forgotten part of them still recognizes the person they loved. A familiar touch awakens fragmented memories of their life together before the world collapsed. The memories feel beautiful, imperfect, and painfully real. The ending returns to the present, revealing that although their humanity is fading, their connection remains. Visual direction: Photorealistic cinema, natural skin textures, anamorphic lens character, subtle film grain, shallow depth of field, realistic handheld movement, soft lens flares, volumetric light, restrained color grading. Present-day scenes feature desaturated green-grey tones, decaying architecture, drifting dust, and cold overcast light. Memories feature warm amber sunlight, rich natural colors, and gentle halation. Emotional restraint over melodrama. Character continuity — keep identical faces throughout: Man: Early 30s, short messy dark-brown hair, light stubble, silver curb chain, small silver hoop in his left ear. Present-day appearance: shirtless, dark trousers, ashy grey-green skin, mottled veins, cracked lips, dirt-covered shoulders, bruising around the eyes. His irises are clouded milky-grey, with faintly veined pale sclera and small, dull pupils. His gaze remains subtly responsive. Memory appearance: healthy olive skin, dark charcoal crewneck, warm brown eyes, natural smile. Woman: Early 30s, long wavy dirty-blonde hair, small silver nose ring, oversized black knit sweater in the present. Her skin is pale and ashy, with realistic discoloration, faint veins, and dark bruising around the eye sockets. Her irises have a cloudy blue-grey film, glassy surfaces, and muted pupils. Memory appearance: healthy skin, heather-grey T-shirt, clear blue-grey eyes, expressive laughter. Shot-by-shot sequence 0.0–3.0s — The abandoned station A wide cinematic shot of a deserted railway platform swallowed by weeds. A derailed train rests beneath a broken glass canopy. The infected man sits alone on a rusted bench, staring at an old photograph held between dirty fingers. Wind moves loose paper across the platform. He slowly lifts his head as faint footsteps echo in the distance. His cloudy eyes turn toward the sound. Slow dolly-in, cold green-grey palette, atmospheric silence. 3.0–5.5s — Recognition At the far end of the platform, the infected woman emerges through drifting mist. Her oversized black sweater hangs loosely from her shoulders, her dirty-blonde hair moving in the wind. She stops beneath a flickering station light. A medium close-up reveals her pale, filmed-over irises and bruised eye sockets. She studies the distant figure as though trying to remember his face. The camera slowly pushes toward her eyes. 5.5–8.0s — The distance between them Alternating profile shots as they walk toward each other along opposite sides of the platform. Neither moves like a predator. Their steps are uneven and hesitant. The man brushes his fingertips across a rusted metal railing; the woman pauses beside an abandoned child’s bicycle. Their dead eyes remain fixed on one another. A train carriage slides across the foreground, briefly obscuring the frame and revealing them much closer when it clears. 8.0–10.5s — The familiar gesture They stop face-to-face beneath a shattered station clock. The woman raises one trembling hand and touches the silver chain around his neck, recognizing something familiar. He looks down at her hand, then back into her eyes. His expression barely changes, but his breathing catches faintly. Extreme close-up on their fingers touching the chain, followed by a rack focus to their clouded eyes. No snarling, no sudden movements. 10.5–12.0s — Memory begins Her thumb brushes the chain. Match cut to the same gesture in the past: her healthy hand resting against the same silver necklace as the man smiles at her in warm afternoon sunlight. The cold station transforms into a sunlit apartment balcony through a seamless visual transition. The soundtrack remains instrumental, with a delicate piano motif and distant ambient textures. 12.0–14.0s — A life they once had Memory. The couple stand in a tiny apartment kitchen, cooking together. She playfully steals a piece of food from his plate. He pretends to be offended before breaking into a genuine laugh. Their clear brown and blue-grey eyes reflect the warm window light. Imperfect handheld framing, natural movement, lived-in surroundings, intimate realism rather than polished advertising aesthetics. 14.0–16.0s — The last ordinary day Memory. They ride a bicycle together along a quiet countryside road at sunset, the woman seated behind him with her arms around his waist. She rests her cheek against his back and closes her eyes. Tall grass sways beside the road. Golden sunlight spills through the trees. The camera follows from behind, creating the feeling of a cherished moment captured before anyone knew it would become a memory. 16.0–18.0s — The world before A brief, fragmented memory: rain against a bedroom window at night. They lie together beneath a blanket, foreheads touching, smiling quietly. Her fingers trace his cheek while he gently tucks a strand of hair behind her ear. Their eyes are alive, moist, and full of warmth. A lightning flash momentarily illuminates their faces, creating a natural visual bridge to the cold present. 18.0–20.0s — Back to the station Hard match cut to the same framing in the abandoned station. The woman’s hand is still against his cheek, but the skin beneath her fingers is cracked and grey. Her eyes are cloudy and lifeless in color. His dull milky-grey eyes remain locked on hers. The piano falls away, leaving only wind passing through broken windows and the faint metallic groan of the station. 20.0–22.0s — The almost-human moment Extreme close-up, both faces filling the frame. The man slowly raises his hand and rests it against the woman’s cheek. A tear-like trace of moisture catches the light beneath her clouded eye, without restoring its color. She leans into his palm by the smallest amount. Their noses almost touch. Fine cracks, subtle veins, dirt, and realistic skin deterioration remain sharply detailed. No supernatural glow, no sudden transformation. 22.0–24.0s — Still together A wider shot from behind them. They sit side by side on the rusted station bench, hands clasped, facing the abandoned tracks disappearing into the mist. Weeds move in the wind. Far above, the broken canopy lets one narrow beam of golden sunlight fall across their joined hands. The rest of the station remains cold and lifeless. The camera slowly pulls away until they become small figures in the ruined architecture. End on their clasped hands. Fade to black. No title card. Cinematography and performance * Prioritize emotional storytelling through eyes, hands, posture, and subtle facial movement. * Use motivated match cuts that connect identical gestures, compositions, and character positions across past and present. * Maintain consistent facial identity, hairstyle, jewelry, clothing, and body proportions in every shot. * Present-day movement should feel slow, stiff, and physically grounded, with no exaggerated zombie behavior. * Human memories should feel spontaneous, tender, and imperfect, with believable interactions and natural expressions. * Use realistic practical infection makeup, restrained discoloration, cracked skin, mottled veins, and bruised sockets. * Zombie eyes must retain recognizable iris and pupil structure beneath a milky film. Never use glowing eyes, featureless white eyeballs, or black voids. * Build emotion through visual contrast, ambient sound, and restrained pacing rather than dialogue or excessive effects. Negative prompt: Cartoon zombies, anime, video-game graphics, plastic skin, exaggerated monster faces, glowing eyes, solid white eyes, black voids, excessive gore, exposed organs, blood splashes, face inconsistency, changing jewelry, changing eye colors in the present, extra fingers, deformed hands, artificial facial expressions, rubbery movement, oversaturated colors, excessive motion blur, random scene transitions, text, subtitles, logos, watermarks.frozen11d ago6.8K805120161.67%
Only 5 days left to lock in Flova’s discounted rates! Get up to 80% off Seedance 2.5 and lock in your pricing for 365 days. • 480p: $0.022/sec • 720p: $0.044/sec • Image 2.5: $0.011/image • Wan 3.0 Prime: $0.022/sec Flova Savings Month ends October 20. Don’t miss out! https://t.co/FoDpVu0wMSfrozen7d ago6.6K100120021.85%
AI video is getting scary good. Higgsfield Genjutsu is built different. https://t.co/TfLddtYJR7frozen6d ago6.5K881171111.81%
Simulating 30 buyers in 30 minutes instead of running an 8-week focus group is a pretty wild shift. Curious to see what they said about the new iPhones.frozen29d ago6.5K72113031.37%
A samurai wrecking everything around them, with dialogue that actually came out right. The problem was the background music and the sound effects. Regenerating to fix those means rolling the dice on the dialogue again. @MireloAI SFX 2.0 kept the dialogue, pulled the music and swapped in new effects. Mirelo is cooking.frozen8d ago6.1K73030051.78%
Told the AI to turn my girlfriend into a love song, and it somehow made our little story sound cinematic. ❤️🎶 Wrote the lyrics, generated the music, and had a full track in minutes. No complicated setup, no download needed. Who would you turn into a song? Drop it below and I’ll make one. 👇frozen15d ago6.0K84020031.77%
Generated with Nano Banana Pro on Gemini app. Prompt: Studio portrait of one smiling person holding a Cheetos Crunchy orange chips bag close to the camera, both the bag and the person’s head in the foreground. Captured with an ultra-wide fisheye lens for a playful, youthful, energetic feel. The model wears a glossy bright orange puffer jacket over a white t-shirt. Background in a vibrant orange gradient with light yellow accents, bold and fun. Soft cinematic lighting, juicy highlights, slight saturation boost, modern snack-ad editorial style. Expressive smile, fun street-commercial aesthetic.frozen19d ago5.9K81331182.09%
Grok can now do more than just talk about crypto markets. 👀 With @senpi_ai, you can connect Grok to live market data, explore trading strategies, and put them into action with your approval. The best part? No coding or API setup required. Check out the demo to see how it works in practice. AI is slowly moving from giving insights to actually getting things done. Not financial advice.frozen10d ago5.9K98212001.90%
EAGLE’s journey from EAGLE-1 to EAGLE-4 shows how research can evolve into real-world infrastructure. Led by Hongyang Zhang, the work on speculative decoding has gained adoption across inference ecosystems like vLLM, SGLang, and TensorRT-LLM. EAGLE-4 is the next chapter in that journey, pushing inference acceleration further.2d ago5.9K80110001.54%
When AI takes road rage airborne. Generated with Seedance 2.5 on @frameo_ai #MadeWithFrameo https://t.co/4BxQYlm7KBfrozen4d ago5.9K90220021.94%
Say no more, my computer is already starting to heat up, but I still really want to take a couple of turns around. #HelixWorld #RealTime #worldmodel https://t.co/O3oHADt3kkfrozen6d ago5.9K6918011.35%
Two photos, and suddenly it feels like a full music video. @noiz_ai turns a simple idea into something cinematic, creative, and full of life. AI is blurring the line between imagination and reality. #noizai https://t.co/ePc2lMLlM2frozen13d ago5.4K86122022.05%
Doomy showing off his skills. https://t.co/f6M13XYb7L3d ago5.4K8459011.85%
TikTok selling is going autonomous. @WithAmplify learns your brand from your website, creates an AI influencer, and posts TikToks in your voice every day. First month is free. Your whole TikTok content team, automated. Try it: https://t.co/Ffj8VPdRyZfrozen20d ago5.2K43217041.26%
The next step for personal AI isn’t just doing more for you. It’s understanding what you’re looking at. NewEyes AI brings the idea of a personal agent with eyes into focus. The real challenge is knowing what “this” means in the moment, not just responding to a promptfrozen10d ago5.0K551150141.70%
Somebody give Drama 3 to every game jam team. Fully voiced weekend games are about to be a thing.frozen9d ago4.9K42070201.41%
Just joined the waitlist and claimed my Creator DNA on @theattractorai. Reserving my creator name early makes it feel a little more personal and gives you a head start on building your identity before launch. If you’re an AI creator, UGC creator, editor, or clipper, definitely worth checking out. Only 5,000 founding creator spots are available. Claim your Creator DNA and join the waitlist through my link 👇frozen21d ago4.9K451150131.52%
Building an entire product from one idea is crazy. AI is changing how fast we can go from concept to reality.frozen15d ago4.7K5942210162.96%
This is where agentic creation gets interesting. Hedra MCP + CLI bringing image, video, and world models directly into your workflow is a big step toward building with vision models, not just using them. Definitely worth exploring.frozen22d ago4.6K71110021.84%
Everyone said chips are too hard for startups. Sabi: designed one, got it made, raised $50M. Hard things matter again.frozen13d ago4.2K451100191.77%
Most agent bugs I’ve chased were found by a user, weeks late. DigitalOcean’s Signals and Simulations, now in public preview, take the session that went wrong, turn it into a test, and replay your approved fix against it before you ship. That’s the loop I’ve been duct-taping together myself.frozen10d ago4.1K560140192.15%
This is what AI app building should look like. Frontend, backend, database, and you actually own the code. Myndlab looks promising.frozen21d ago4.1K39090171.59%
Last month I spent 3 days trying to get 6 people to agree on one weekend. Spreadsheet, poll, voice notes, the whole thing. Wabi would've done it in one message.  Dates, itinerary, flights. I'd have just shown up. https://t.co/ctdpis3RbQfrozen3d ago3.6K37040251.82%
AI is changing the economics of marketing. When you can scale content with AI at a fraction of the cost, the traditional model starts looking very different. The companies that adapt early will have a serious advantage.frozen15d ago3.6K510210152.42%
Gojiberry just shook up the entire data broker industry. https://t.co/DCaOIO2WrGfrozen14d ago3.4K6528012.23%
Marketing teams are getting a serious AI upgrade. From brand research and campaign planning to creating ads and analyzing performance, GooseworksAI brings the entire workflow into one place through Dots. Worth exploring for anyone building a brand with AI.frozen9d ago3.1K49216022.22%
MiMo 2.6 Pro on Inworld Inference: cached input is $0.0036 per million, uncached is $0.425. Long chats keep processing the same history every turn, so that gap adds up quick.frozen8d ago2.9K470100252.80%
Viral Halloween ads used to get booked in August. Now you can make one in minutes. Pick one of 93 Halloween presets. Add your product. It's live in minutes. Grab 50% off BuzzHooks now.7h ago2.9K4917012.03%
Your ads did their job. The lead just never booked, or the reply came too late. iClosed Unified Inbox captures those missed leads and gets replies out in under 5 minutes. That’s how you stop losing demand you’ve already paid for. @iclosedsaas https://t.co/LzGgpLf7Qh5h ago4063101001714.29%
These two look like they belong in their own animated series. 🏀✨ With @flovaai’s High-End Character Face Design Skill, you can bring original characters to life with detailed facial close-ups and full-body designs. Watch from 0:18 to see how it’s done. https://t.co/K6uLfTMAeE4h ago208404003.85%