Google's open-source, JAX-native post-training library Tunix now ships an agentic RL trainer built around asynchronous rollouts and a decoupled producer-consumer pipeline, aiming to eliminate idle TPU…
Microsoft released Agent Lightning v1.0, an MIT-licensed open-source reinforcement learning framework that trains AI agents without modifying agent code, available on GitHub. Using the framework, a Qw…
Microsoft released Agent Lightning v1.0, an open-source reinforcement learning framework that lets developers train AI agents without altering their production setup, on August 17. The framework, whic…
Hugging Face published an August 19 article by Old Dominion University graduate researcher Michael Evans introducing Bedrock-RL, an open-source framework that makes Minecraft experiments repeatable fo…
AMD's ROCm blog published a walkthrough for running verl's asynchronous reinforcement learning examples on AMD Instinct MI355X GPUs, covering GRPO on Qwen2.5-VL-7B-Instruct with the Geometry3k dataset…
NVIDIA's NeMo team released Molt, a PyTorch-native agentic reinforcement learning framework designed to be compact enough for researchers and AI coding assistants to grasp, with roughly 8.6K lines of …
Anthropic reported that more than 80% of the code merged into its codebase was authored by Claude, with engineers merging 8x more code per day than in 2024, but an independent test by a researcher fou…
EasyOPD, a new framework built on the verl platform, simplifies AI model training by integrating on-policy distillation (OPD) methods into a unified system, enabling faster and more efficient model ev…