Show HN: A decent text-to-image model trained for ~$20K (Apache 2.0 Weights)
Fonte: Show HN ·
My brother and I run a 2-person lab that trains our own generative video models from scratch. The ultimate goal is a set of creative tools that make it easier for anyone to make animation. Video models today are too slow/expensive and lack sufficient controls. Creativity is an iterative process. You make, you tweak, you get better. In our opinion, they're not usable yet for storytelling. When we set out to build Linum v3 (our next video model) earlier this year, the primary goal was to bring the cost envelope of inference down significantly. If our model is fast and cheap, it's the first step
Domande frequenti
Che cos'è Show HN: A decent text-to-image model trained for ~$20K (Apache 2.0 Weights)?
Show HN: A decent text-to-image model trained for ~$20K (Apache 2.0 Weights) is a strumento discovery curated from Show HN. My brother and I run a 2-person lab that trains our own generative video models from scratch. The ultimate goal is a set of creative tools that make it easier for anyone to make an
Come posso accedervi o provarlo?
You can experience this content directly on its original host (Show HN) by clicking the "Open Original" button.
Da dove proviene questo contenuto?
Curated from Show HN by the OnSuggest team to surface the most interesting and serendipitous content on the web.
Altri simili
Tutte le scoperte in Strumenti per sviluppatori →Show HN: AegisKernel – Zero-trust POSIX runtime for AI agents in <0.016ms
Show HN: A crawler found 2,900 TV apps on GitHub
An app store for Android TV that installs open-source TV apps from GitHub releases - wingedpanther/openshelf
Show HN: Hacker Living – HN's most recommended places
HackerTV and HackerTube inspired me to explore another corner of HN recommendations: places worth visiting and typefaces people enjoy using. I built: https://hacker.living - a map and font collections with links back to the original comment. I built this entirely using Codex, including the entire data pipeline and workers: extraction, transformation, and recommendations capture. It's still an early stage preview, so some details need verification. Thanks for the inspiration! Code: https://github.com/grandimam/hacker.living Comments URL: https://news.ycombinator.com/item?id=50395134 Points: 1 #
Show HN: Ru-Jev – an open model that scores caller-supplied options on CPU (MIT)
Text Classification model with 0.3B parameters. Download, explore files and run it on Hugging Face.
Show HN: Wrapped – Your neighborhood's 311 story
Explore 2026 so far neighborhood 311 reports across US cities. Enter a ZIP for animated maps, top complaints and a personalized share card. No sign-up.
Show HN: Thesis, checks every SEC filing against why you own a stock
Write down why you own each stock. Thesis checks every filing, earnings release and headline against your thesis, and alerts you when one touches it.