abidlabs

If this doesn't put the nail in the coffin on the idea that we need closed-source models for the good of cybersecurity, I don't know what will

Yes! https://huggingface.co/changelog/agent-trace-viewer

Interesting to see Anthropic now downplaying the new vulnerabilities that Mythos discovered:

We reviewed a demonstration of this specific technique being used to identify a small number of previously known, minor vulnerabilities. These vulnerabilities all appear relatively simple, and we have found that other publicly-available models are able to discover them as well without requiring a bypass

Would you be interested in an RSS feed for: https://huggingface.co/papers/trending?

You could subscribe to daily papers at the top here: https://huggingface.co/papers/trending

Hugging Face Buckets also recently added support for mounting Buckets as a filesystem: https://huggingface.co/changelog/hf-mount

The most interesting thing is that charges can be potentially much less if you have duplicated data in your bucket

Use Hugging Face inference providers for this: https://huggingface.co/docs/inference-providers/en/index

https://archive.is/6DwJL

Test it with a web UI: https://huggingface.co/spaces/abidlabs/openai-gpt-oss-120b-t...

You could use Hugging Face's Inference API (which supports all of these API providers) directly making it easier to switch between them, e.g. look at the panel on the right on: https://huggingface.co/openai/whisper-large-v3

Love the recipes and docs here, kudos on the launch!

Official Gradio demo is here: https://huggingface.co/spaces/stabilityai/stable-fast-3d

(b) growth is still the right metric imo. for example, lots of open-source libraries (which by definition are available to use for free) measure and optimize for growth

We have ~220 total team members across all roles

Lucky to be part of this company and seen this strategy work close up. One thing I'll add is this "decentralized" approach applies to all Hugging Face teams, not just the developer advocacy team. Just to give an example, there's no central comms team at Hugging Face, every team (usually the engineers who work on the product or features) does their own comms across the channels they think work best. That means there's lots of experimentation and most of our hires tend to be generalists who are comfortable wearing many hats.

Hugging Face Paper Page and Discussion: https://huggingface.co/papers/2404.14219

Yes you’re right thanks, I should have disclaimed earlier, but AK is a member of the team I lead at Hugging Face.

There are have been many, many services that have tried to automate paper selection based on these heuristics. None of them have had the staying power of AK's account. As someone with a PhD in machine learning from Stanford, I can attest AK's taste is quite good.

Is there a Gradio demo?

Is it possible to provide Gradio apps as an option?

What's the underlying technology that allows this to happen? Is it Pyodide-based? And does this support GPU-based models?

What were your test questions?

Please add support for open-source LLMs from Hugging Face :)

Is the trained model available on Hugging Face?

A better example would have been the RWKV [1]

[1] https://github.com/BlinkDL/RWKV-LM

That's actually not true: https://github.com/BlinkDL/RWKV-LM

Just a matter of time, until plugins are supported!

due to multimodality it may actually be more arcane to use.

Can you elaborate?

Follow up Guide that explains how to create your own tools: https://huggingface.co/docs/transformers/custom_tools