7arXiv cs.CL (Computation and Language)·
OpenForgeRL: Open-source framework for end-to-end RL training of harness-native AI agents
OpenForgeRL is an open-source framework that enables end-to-end reinforcement learning training of AI agents operating within real inference harnesses (e.g., Claude Code, Codex, OpenClaw) and diverse environments. It uses a lightweight proxy to record harness model calls as training data for standard RL codebases like veRL, plus a Kubernetes orchestrator for scalable rollouts in isolated containers. Trained agents (OpenForgeClaw and OpenForgeGUI) achieve competitive results on benchmarks including ClawEval, OSWorld-Verified, Online-Mind2Web, and WebVoyager, matching or surpassing models several times larger in GUI tasks. The work also analyzes how harness choice and RL shape agent behavior, finding meaningful variation in learnability across harnesses.
Related guides (4)
Related events (8)
Open source community rallies around OpenEnv for agentic reinforcement learning
A Hugging Face blog post announces community backing for OpenEnv, an open-source environment framework targeting agentic reinforcement learning. The post highlights growing open-source momentum around training infrastructure for RL-based agents. This signals a potential consolidation point in the fragmented landscape of agentic RL tooling.
Unlocking Agentic RL Training for GPT-OSS: A Practical Retrospective
A Hugging Face blog post authored by LinkedIn describes practical lessons from implementing reinforcement learning training for agentic open-source GPT-class models. The retrospective covers engineering and algorithmic challenges encountered when applying RL to agentic workflows. As a tier-2 source with no body content available, the depth and specific findings cannot be fully assessed, but the topic sits at the intersection of agentic systems and RLHF/RL training pipelines.
Putting RL back in RLHF: RLOO Implementation on Hugging Face
Hugging Face published a blog post introducing RLOO (REINFORCE Leave-One-Out), a reinforcement learning algorithm aimed at making the RL component of RLHF more practical and effective. The post discusses implementation details and motivations for revisiting pure RL-based fine-tuning approaches within the TRL library. This represents a technical contribution to the alignment and RLHF tooling ecosystem, offering an alternative to PPO-based RLHF pipelines.
OpenHarness: Open-source agent harness with built-in personal agent (Ohmo)
HKUDS has released OpenHarness, an open-source agent harness framework in Python that includes a built-in personal agent called Ohmo. The project has accumulated nearly 15,000 GitHub stars, suggesting meaningful community traction. It appears to be a general-purpose scaffolding layer for building and evaluating AI agents.
OpenRAIL: Towards open and responsible AI licensing frameworks
This Hugging Face blog post introduces the OpenRAIL licensing framework, designed to enable open access to AI models and datasets while embedding use-based restrictions to prevent harmful applications. The framework attempts to balance openness with responsibility by attaching behavioral conditions to model redistribution and use. It represents an early effort to create AI-specific licensing that goes beyond traditional open-source software licenses.
OpenPipe ART: Agent Reinforcement Trainer for Multi-Step Agents via GRPO
OpenPipe has released ART (Agent Reinforcement Trainer), an open-source Python library for training multi-step agents on real-world tasks using GRPO (Group Relative Policy Optimization). The framework supports multiple model families including Qwen3, GPT-OSS, and Llama. With nearly 10k GitHub stars and 66 gained today, it is gaining notable community traction as a practical RL fine-tuning tool for agentic workflows.
OpenAI Releases RL-Teacher: Open-Source Human Feedback Interface for RL
OpenAI released RL-Teacher, an open-source implementation of an interface for training AI systems using occasional human feedback instead of hand-crafted reward functions. The tool implements a technique developed as a step toward safer AI systems and is applicable to reinforcement learning problems where reward specification is difficult. This represents an early public release of human-in-the-loop RL tooling from OpenAI.
We Got Claude to Fine-Tune an Open Source LLM
Hugging Face demonstrates using Claude (Anthropic's model) as an orchestrating agent to autonomously fine-tune an open-source LLM, showcasing an agentic workflow for model training. The post illustrates how a frontier model can handle the end-to-end process of dataset preparation, training configuration, and execution for a smaller open-weights model. This represents a practical example of AI-assisted ML engineering and agent-tool ecosystem development.