OpenEnv is a tool for creating an agentic execution environment like terminals, browsers, or anything an agent can interact with. And today, we’re excited to announce that OpenEnv is becoming even more open, to make the future of training agents open source. Starting today, OpenEnv will be coordinated by a committee that so far includes Meta-PyTorch, Reflection, Unsloth, Modal, Prime Intellect, Nvidia, Mercor, Fleet AI, and Hugging Face. OpenEnv now lives at huggingface/OpenEnv OpenEnv project is supported and adopted by […]

If you have a GitHub repository and you have GitHub Actions enabled, you probably use GitHub-hosted runners for CI. That is the default for many projects because it is simple: add a workflow, write runs-on: ubuntu-latest, and GitHub gives you a machine. That default is convenient, but it also has limits. GitHub Actions can be slow or down for maintenance, the hosted machines are generic, and GPU access is not something most open-source projects can just turn on. For Trackio, […]

An agent built a 3D Paris gallery from two Hugging Face Spaces. I asked a coding agent to build a beautiful website showcasing the monuments of Paris as 3D Gaussian splats. I never opened an image generator. I never touched a 3D reconstruction tool. The agent produced every asset (the images and the 3D splats) by calling two Hugging Face Spaces directly, then wired them into a cinematic viewer. Here’s the result, live as a static Space: 👉 mishig/monuments-de-paris This […]

All co-authors listed below Today, we are releasing North Mini Code, a 30B-parameter Mixture-of-Experts model with 3B active parameters with powerful agentic coding capabilities, available on Hugging Face under the Apache 2.0 license. North Mini Code is the first model in Cohere’s new family of models, and is specifically designed and trained for agentic software engineering tasks. Figure 1: North Mini Code’s performance in agentic coding tasks and complex code generation benchmarks, compared to leading open-source models of similar size. […]

In the first part of this series “Profiling in PyTorch”, we used torch.add(torch.matmul(x, w), b) to learn how to read PyTorch profiler traces. We also discussed several other topics that came our way – the CPU dispatch chain, launch overhead, the difference between an overhead-bound and a compute-bound regime, and some internals of torch.compile. In the second iteration (this blog post), we climb one rung up the ladder. We replace the hand-written matmul-add pair with an nn.Linear (with bias=True). This […]

💻 Code: https://github.com/allenai/olmo-eval While you’re building an LLM, you evaluate it over and over across many interventions. Every adjustment to its data, architecture, or hyperparameters — and every step up in scale — sends you back through the same loop: adding or reconfiguring benchmarks, re-running them on each new model checkpoint, noting the results, and checking whether something that helped in a small experiment still holds up on the full training run. Most evaluation tools aren’t designed for this—they’re either […]