Mishig ᠮᠢᠰᠾᠢᠭ
4,298 posts
Mishig ᠮᠢᠰᠾᠢᠭ
@mishig25
artificial intelligence @huggingface
- How do you draw an owl: 1. draw some circles 2. click "Diffuse the rest" We have made the greatest thing since MS Paint (using #stablediffusion): huggingface.co/spaces/hugging… Looking forward to seeing what you'll generate with it 😊
- As exact as it gets
what the fuck is "attention," exactly - text-to-3D text -> 2D image -> 3D point cloud Step 1: describe what you want -> Step 2: img generation with a diffusion model -> Step 3: 3D point cloud generation using a second diffusion model using the img from step 2 gr.demo: huggingface.co/spaces/osansev…
00:00 - Microsoft upgraded its search. Google upgraded its search. Github upgraded its search. Now, it is our turn. Let us begin: hf.co/search
00:00 - Shipped. For instance, you can see that gpt-oss-120b is 196 GB right from the “Files” tab
The feature I most want from @huggingface is a prominent "total repository size" figure somewhere near their "files" browsing interface If they have that already I've managed not to find it so far! - SSI strategy of not releasing a product is probably a good one. The minute one releases a product, one will be dragged into so fierce competition with OAI, gemini, ... that the original goal will be forgotten. Maybe it would have been wiser for Anthropic to never release Claude
- It is cool to be capable. It is cool to know shit. That's why the HF team is open-sourcing not just the model, but the training code and datasets too. Learn. Build. Make it your own.
- ML runs everywhere now because of 2 recent trends: * 𝚓𝚊𝚟𝚊𝚜𝚌𝚛𝚒𝚙𝚝-𝚏𝚒𝚌𝚊𝚝𝚒𝚘𝚗 of ML (transformers.js, huggingface.js) for running on browsers * 𝚌𝚙𝚙-𝚏𝚒𝚌𝚊𝚝𝚒𝚘𝚗 of ML (llama.cpp, bloomz.cpp, whisper.cpp) for running on embedded devices
- As the field of ML-for-3D grows, we've added 3D model viewer for 🤗Hub datasets For example, checkout this Cybertruck from @allen_ai 𝗢𝗯𝗷𝗮𝘃𝗲𝗿𝘀𝗲 dataset: huggingface.co/datasets/allen…
00:00 - Pose 3D mode & render with Stable Diffusion (ControlNet) (forget about ray-tracing, path-tracing, or any kind of tracing in that respect #FutureIsNow) try gr.demo: huggingface.co/spaces/diffuse…
00:00 - text-to-image-editing words -> mask -> replacement describe what you want -> automatic mask generation with CLIPSeg -> replace mask with Stable Diffusion generation gr.demo: huggingface.co/spaces/nielsr/…
00:00 - Not sure if RLHF is the future, but it is definitely the present. As such, you can find RLHF datasets on the Hub 1. @OpenAI / WebGPT Comparisons huggingface.co/datasets/opena…
00:00
