• Home
  • Technology
  • Gaming
  • Entertainment
  • World & Business
  • Science
  • Sports
  • AI
HomeTechnologyGamingEntertainmentWorld & BusinessScienceSportsAI
Technology
Announcement

RL environments reportedly now have a home on the Hugging Face Hub

In a post announcing the change, Ben Burtenshaw said reinforcement learning environments should work more like datasets on the Hub, which he said already stores, versions, gates, and previews data.

Ben BurtenshawBB
Adithya S KAS
Hugging FaceHF
3 Sources, 5d ago, first seen 5d ago

TLDR

A post from Ben Burtenshaw said RL environments now have a home on the Hugging Face Hub, framing them as something that should be handled like datasets. He said the Hub already stores, versions, gates, and previews data. A separate user post said developers can discover environments from OpenEnv, Verifiers, Harbor, and NeMo Gym there, and Hugging Face later reshared that post.

Combined views

51.4K

3 Sources, first seen 5d ago

270 likes17 comments193 saves37 reposts

Combined views

51.4K

3 Sources, first seen 5d ago

270 likes17 comments193 saves37 reposts

Ben Burtenshaw said in an announcement post that RL environments now have a home on the Hugging Face Hub.

Featured Source

In that post, Burtenshaw described a fragmented setup across reinforcement learning frameworks, writing that developers have relied on separate ways to find environments, including custom hubs, runtime registries, and GitHub lists with custom loaders. He argued that this leaves environments published for one framework harder for users of others to load.

Burtenshaw framed environments as something that should live on the Hub much like datasets. He wrote that environments are made from tasks, tests, containers, and reward functions, and said the Hub already stores, versions, gates, and previews data for millions of people, so it does not need a separate system for this category.

A separate post from @adithya_s_k said developers can now discover environments across OpenEnv, Verifiers, Harbor, NeMo Gym, and more on the Hub. That cross-project discovery list appears in the user post, not in the supplied announcement text from Burtenshaw.

Hugging Face later reshared that post. Based on the supplied evidence, that reshare shows the company account acknowledged the post, but it does not independently verify the technical details listed there.

Sentiment

Positive——Negative

Summary

Not enough discussion yet.

No sentiment analysis available yet.

Sentiment

Positive——Negative

Summary

Not enough discussion yet.

No sentiment analysis available yet.

Related

Hugging Face reposted Victor M @victormustar's Microduck claim

Victor M @victormustar wrote that Hugging Face is making robotics on the Hugging Face Hub easier to use and that users will "soon" be able to send any community-made policy to a Microduck. Hugging Face reposted that message on X.

HuggingChat displays an “ML Intern” label in its interface

Hugging Face linked to the hf.co/chat?mode=ml-intern page in an Oct. 1, 2026 X post about MCPs, but the supplied chat-page capture does not explain exactly how “ML Intern” works.

Gallouédec posted “Next stop: 1M/week,” and Hugging Face reposted it

Quentin Gallouédec wrote that 1 million trainings a week is the next goal and framed post-training as a step toward a less centralized AI ecosystem. Hugging Face later reposted the post.

3 Sources

Ben Burtenshaw@ben_burtenshawRL environments now have a home on the @huggingface Hub. Every RL framework has its own way to find environments. Custom hubs, runtime registries, GitHub lists with a custom loader. If you publish for one framework, users of the others can’t load it. We think that is wrong. Environments should be just like datasets because it’s made from tasks, tests, containers and reward functions. The Hub already stores, versions, gates and previews data for millions of people so it doesn’t need a second system. Here’s how we’re shipping environments on the hub: http://huggingface.co/blog/rl-environments5d
Adithya S K@adithya_s_kThis is hugeeee for open-source RL environments. You can now discover environments across OpenEnv, Verifiers, Harbor, NeMo Gym, and more, all on the @huggingface Hub. A common home for RL environments is long overdue. Excited for what’s coming next !5d
Hugging Face@huggingfaceRT @adithya_s_k: This is hugeeee for open-source RL environments. You can now discover environments across OpenEnv, Verifiers, Harbor, NeM…5d
    • Home
    • Technology
    • Gaming
    • Entertainment
    • World & Business
    • Science
    • Sports
    • AI
    Hugging Face

    3 Sources

    Ben Burtenshaw@ben_burtenshawRL environments now have a home on the @huggingface Hub. Every RL framework has its own way to find environments. Custom hubs, runtime registries, GitHub lists with a custom loader. If you publish for one framework, users of the others can’t load it. We think that is wrong. Environments should be just like datasets because it’s made from tasks, tests, containers and reward functions. The Hub already stores, versions, gates and previews data for millions of people so it doesn’t need a second system. Here’s how we’re shipping environments on the hub: http://huggingface.co/blog/rl-environments5d
    Adithya S K@adithya_s_kThis is hugeeee for open-source RL environments. You can now discover environments across OpenEnv, Verifiers, Harbor, NeMo Gym, and more, all on the @huggingface Hub. A common home for RL environments is long overdue. Excited for what’s coming next !5d
    Hugging Face@huggingfaceRT @adithya_s_k: This is hugeeee for open-source RL environments. You can now discover environments across OpenEnv, Verifiers, Harbor, NeM…5d
    Today's Rank

    —

    Not ranked yet

    Today's Rank

    —

    Not ranked yet