Changelog

The tensor viewer now loads much faster on large sharded models. When you open a model's model.safetensors.index.json, the Hub sends the headers of all shards in a single request, so your browser no longer fetches each shard one by one. For example, DeepSeek-V3.2 now loads in 0.45s instead of 9.2s, GLM-5.3 in 0.57s instead of 4.0s, and Qwen2.5-72B-Instruct in 0.27s instead of 1.2s.

Fine-grained access tokens and service account tokens can now be granted a new contributor permission. A token with it can create repositories in your account or in the organizations you select, and gets full write access to the repositories it created itself. It can't write to any other repository, even ones you or your other tokens created.

This is useful for CI/CD pipelines, training jobs and agents that publish new checkpoints or datasets: they no longer need write access to every repository in the namespace. It works like the contribute-repos scope already available to OAuth apps.

Decision models now have their own decision-model tag, with a dedicated icon on model pages and a filter in the Misc section of the Models page. GGUF decision models are tagged automatically from their llama.cpp metadata, so new conversions are picked up as soon as they're uploaded.

You can now add your P(doom) to your Hugging Face profile: your estimated probability of AI causing an existential catastrophe. Set it from your profile.

Your answer also contributes to a community survey on AI risk perception. Only aggregated, anonymized results will be published, in two weeks.

The job page has a new Resources panel next to the logs. While your job runs, it shows live CPU, RAM, GPU and VRAM usage, updated every couple of seconds.