SIMUSOFT @simusoft

The combination of creativity and AI is opening up entirely new forms of storytelling. Excited to see where this goes next. @PixVerse_ #PixVerse #PixVerseOriginals pixverse.ai/en/blog/pixver…

PixVerse @PixVerse_

22 hours ago

Welcome to Mars Landings. Humanity's first off-world resort community. No return shuttles. There never will be. PixVerse Originals S1. @aivideoschool's Mike Vogel built this ensemble sci-fi dark comedy in PixVerse. RT+Follow+Reply "Originals" = 150 Creds & Full Film in DMs (72H

94 78 245 162K 52

🚨A CHINESE LAB HAS JUST HUMILIATED HALF THE VIDEO INDUSTRY. Meituan's LongCat team just dropped LongCat-Video-Avatar 1.5 — a production-ready, open-source framework for audio-driven avatar generation. The input: one photo. One audio file. The output: a fully lip-synced speaking avatar. Most AI avatar tools fall apart after a few seconds. Faces morph, lip sync breaks, identity drifts. LongCat 1.5 was built specifically to solve these problems — optimized for stability, long-form generation, and multi-character interactions. It supports news broadcasting, education, entertainment, singing, e-commerce, and multi-person conversation — in both Chinese and English, across realistic and animated visual styles. It is MIT-licensed. You can use it commercially. What used to require a camera crew, a studio, and post-production editing now runs from a GitHub repo. Version 1.5 also upgrades to Whisper-Large for sharper lip sync and cuts inference down to 8 steps via step distillation — making it significantly faster than before. The tools that charged thousands for this workflow just got a free competitor dropped on them. This is LongCat-Video-Avatar 1.5. It is worth your attention.

Made this anime-style teaser with Seedance 2 inside @ImagineArt_X From ruined Tokyo to that sword draw... the vibes are insane. Tag an anime fan who needs to see this #animefanart #anime

Claude 4.8 + 1 prompt and now anyone can achieve the same results. Prompt below 👇

Here's a teaser of our Mac-1 model. > 6.6B model > runs locally (on any Mac) > requires 7GB RAM (12GB ideal) > can use 487 MacOS native tools > perform multi-tool chained tasks > reasoning: ON > output: ~65 tok/s We built a robust application layer around the model to make UI/UX MacOS native. The "model-focused" SaaS era is here. Stay tuned for more.

CJ Zafir @cjzafir

3 days ago

Our first model Mac-1 6.6B beating 3 giant models. - Haiku 4.5 - GPT 5.4 mini - Gemini 3 flash Running this model on my Macbook M3 24GB. (model takes only 7GB RAM) It searches web, call tools, ask follow-ups, tell jokes, find contacts, search files, write emails, book events,

157 108 2K 713K 1K

Someone built a screen recorder that makes every demo look like an apple ad.. 100% free to try.

NVIDIA just made paying for AI feel optional. Open model, a million tokens of context, free tier with no per-token cost, runs on your own hardware. Entire codebases, whole data rooms, a year of chat logs, all swallowed in one prompt. No chunking, no RAG, no rate limit theater. The closed-AI premium has 90 days to defend itself. Bookmark this and come back. Open beat closed. Again.

shmidt @shmidtqq

2 days ago

x.com/i/article/2063…

32 26 224 393K 579

#Omni

fofr @fofrAI

2 days ago

Start with an image of a building and prompt for a projection mapping with Omni.

5 11 234 46K 119

point and click ⬅️ Made in @ComfyUI + @Photoshop + @AdobeAE with GPT Image 2.0 and Wan 2.2 I2V ✏️🤖

100 Mac Minis quietly became an AI server farm. That is the part people are missing. Hao Wang did not wait for cheaper GPUs. He stacked Apple hardware. And suddenly the economics changed. A $599 Mac Mini does what many developers rent for $200/month. Ollama runs local. Claude Code connects through a local endpoint. The bill drops from hundreds per month to a few dollars in power. This is not a cloud story. It is not a model story. It is a control story. Developers are starting to stop renting intelligence. They are putting it on their desk. Bookmark this before every AI setup starts looking local.

fink @0xfinkus

4 days ago

x.com/i/article/2062…

16 4 89 486K 237

Gopuff + SpaceXAI just made shopping predictive. 13 years of cart data + real-time X signals. => Open the app. => It builds your cart. => Tap once. => Delivered in 15 minutes. Search is getting skipped.

Gopuff @gopuff

5 days ago

Meet Go. Gopuff's AI shopping genius, co-developed with SpaceXAI. Just say what you need. It's already on its way.

572 687 9K 77.1M 2K

SAM ALTMAN HAS A NEW PROBLEM. 🤯 Google just shrunk 31GB of AI memory down to 4GB. The tool is called TurboVec. It uses up to 16x less memory, searches faster than FAISS, runs fully offline, and works on a regular Mac. No expensive GPU cluster. No cloud dependency. No compromise on speed. → 16x lower memory usage → Faster vector search → Works with LangChain & LlamaIndex → 100% open source The race to build bigger AI models is loud. The race to make them dramatically cheaper just got a lot more interesting. Repo: github.com/RyanCodrai/tur…

keyboard_arrow_left Previous keyboard_arrow_right Next

HOW THREE TEENAGERS TO BUILT A $9,700/MONTH PHONE FARM WITH A LOCAL HERMES AGENT three teenagers in china started with 10 cheap android devices running basic scripts under a desk to automate online traffic now they run hundreds of phones managed by a local hermes agent to execute B2B marketing and sales campaigns on autopilot the agent acts as the main system gateway, managing files, writing automation rules, and coordinating phone actions without tokens the setup is highly optimized: - deploy a local ollama endpoint to run a quantized model for free - sync the physical phone racks to a group control client for action dispatch - configure hermes to write custom skills for content distribution and leads routing - use dynamic ip rotation to bypass platform anti-bot filters this turns cheap local hardware and silicon into a highly efficient digital employee node if you want to build an automated farm or launch a B2B business using local compute ↓

marfin @marfinxx

2 days ago

x.com/i/article/2062…

4 0 31 35K 102

23-YEAR-OLD GAMER VIBE CODED A MOBILE APP IN 14 DAYS. 12,000 DOWNLOADS IN 50 DAYS. NOW IT MAKES $20,000/MO AND WON A HYPER-COMPETITIVE HACKATHON he didn’t try to invent a complex tech ecosystem. he built a utility app for one obsessed niche: people in the US looking for class action lawsuit payouts they are already eligible for. it taps into a core human desire -> making free money with zero friction the product is simple. onboarding, a list of active lawsuits (like NBA Top Shot), requirements, and an automated PDF form generator. that is it. the hard part was never the idea. it was needing a team of mobile developers, backenders, and designers to ship it fast claude code changes that. he had no computer science background, but by feeding claude a clear JSON text document of his data structure and competitor screenshots, he vibe coded the entire app cross-platform using expo, next.js, and supabase in two weeks on his own the monetization is aggressive from day one. 90% of users only see the onboarding before the paywall hits. he offers weekly and yearly subscriptions, fine-tuned via revenuecat to push the high-lifetime-value yearly option. people gladly pay for an app when they know it directly unlocks cash payouts most people still think apps need dozens of features to succeed. he proved the opposite. you only need 1 to 3 core features. the onboarding does 100% of the heavy lifting. if it invokes emotion, looks personalized, and shows scientific proof, it converts like crazy distribution wasn't about burning millions on broad ads. he partnered with a hyper-targeted niche content creator to drive thousands of organic UGC downloads. then he took those viral, entertaining videos and plugged them into facebook ads to scale predictably the era of over-engineered software is dead. vibe coding just made it cheap enough for one optimistic kid with a laptop to out-compete 55,000 people and build a cash-flowing machine from his bedroom

Ridark @ridark_eth

24 hours ago

x.com/i/article/2063…

25 15 135 99K 295

This video alone proves photorealism has been achieved in video games 🔥

GPT-5.5 in Codex made an Electric Toothbrush in ForgeCAD. Incredible!

keyboard_arrow_left Previous keyboard_arrow_right Next

We’re hiring creators all over the world to tell stories humanity needs to hear. Stories our children need to watch. - AI video film makers - traditional script writers - musicians traditional or AI. We’re aggressively hiring and the opportunities are endless. 👎

Google Omni might be too powerful 🫥

#Avatars

Vyom 👾 @HelloVyom

2 weeks ago

stumbled onto something genuinely impressive today anam.ai - real-time interactive AI avatars with 180ms latency let me break down why I think this is one of the more underrated AI products right now 👇 most AI agents are either a chat box or a robotic voice.

29 86 1K 264K 2K