prime-rl 0.9.0 Adds Local Run Dashboard and Parallel Agent Evals

prime-rl 0.9.0 introduces a local run dashboard with logs, metrics, rollout and config viewers, and adds support for running agent evaluations in parallel with SFT training by deploying an inference server that reuses weight broadcast infrastructure.

2026-08-26 ~ 2026-08-26 · 2 related posts

• prime-RL adds Local Run Dashboard with logs and metrics viewer — samsja19 · 2026-08-26
• prime-RL supports online Agentic Evals alongside SFT training — samsja19 · 2026-08-26