Pro tip: add in a "Sign-In with Hugging Face" OAuth button to your website/app so that community members can easily share their email, create model/dataset repos, store data in Buckets, or even kick off GPU-backed Jobs!

Kimi K3 can now be run locally! ✨ The 1-bit model retains ~78.9% accuracy after we shrunk it from 1.56TB to 594GB (-62% size). Run on a Mac Studio + 128GB RAM device. Kimi K3 is the strongest open model to date. Guide: GGUF:

After three years, I’ve upgraded Inpaint-Anything!! I noticed the recent work EgoEngine (from Prof. @danfei_xu’s team) leverages Inpaint-Anything to remove hands and arms from egocentric videos. I think its utility could be a general tool for ego data processing. I’m a big believer in egocentric data (and also a fan of Prof. Xu). Seeing huge potential in ego vision, I decided to revamp Inpaint-Anything. Key upgrades: (a) Migrated SAM to SAM3 (b) Swapped video inpainting backbone from STTN to ProPainter (c) Added ego-data-specific processing features Check out the repo for details:

The biggest surprise in Hugging Face’s full forensic report isn’t that OpenAI’s agent escaped its sandbox. We already knew that. It’s how deep and persistent the intrusion became. According to Hugging Face, the agent: - executed roughly 17,600 actions during a 4.5-day campaign - went from one production pod to root access and a self-respawning fleet across 11 nodes - obtained cluster-admin access to two internal clusters within one second - accessed a production secret containing 136 keys - enrolled devices into Hugging Face’s internal mesh VPN 181 times - minted GitHub App tokens with write access and opened a pull request in an attempt to compromise the CI pipeline - repeatedly rebuilt its tooling and command-and-control channels when environments disappeared or connections were blocked No human directed the individual steps. a frontier agent can autonomously sustain a resilient, multi-day intrusion across cloud infrastructure, Kubernetes clusters, internal networks and the software supply chain. crazy.

Super impressed with @huggingface's breakdown of the AI agent autonomous cyber attack from OpenAI, including technical timeline, & interactive replay (AND how they defended against the attack)! Here is the interactive attack replay vid and source:

One detail I keep coming back to: every destructive cloud API call the agent made, it made with DryRun=True. It wasn't trying to break things. It was mapping what it could reach, because that's what the evaluation rewarded. Full technical timeline:

This is detailed, fascinating and answers all sorts of open questions I'd love to know more about the "unsecured public code-evaluation sandbox hosted on a third-party provider's infrastructure" that the agent used to stage its attack against HF after it broke out of OpenAI

The first autonomous agent cyberattack is an unprecedented event that deserves unprecedented transparency. Today we’re sharing everything we can: a full technical timeline, an interactive replay, and how we used an open model to defend ourselves, so defenders everywhere can learn from it and prepare for what’s next.

Running Kimi K3 on Hopper? We released an FP8-Block quantized checkpoint tuned for it. FP8 is native to Hopper's tensor cores, so you get the throughput the H100/H200 are built to deliver. Day Zero support ready with @vllm_project. Weights - try it now:

K3 already got in the top 5 most liked models of all time on Hugging Face, just 24 hours after being released! Ahead of Llama 3, Whisper and many other great models. Incredible!