TRL and OpenEnv now let you train coding agents end to end with opencode
SergioPaniego · x · 2026-07-24
- The post says you can now train your own coding agents with TRL + OpenEnv, starting from opencode.
- The new setup adds end-to-end support for training agent harnesses: TRL launches the agent inside an OpenEnv session, captures its trace, reconstructs training samples, and trains with AsyncGRPO.
- OpenEnv acts as the OpenCode harness environment plus a transparent proxy that records each turn’s token IDs and logprobs.
- The key idea is to train the actual agent loop “as is,” so the policy learns from the exact tokens it produced while using tools.
- They also mention a self-contained example using a local subprocess sandbox and DeepCoder problems, validated on Qwen3-8B.
Related event: TRL and OpenEnv Enable Custom Agent Harness Training(4 posts)→
More from coding & agent
- DeepMind paper: LLMs can't sample randomly on their own — pass random numbers from code instead — rohanpaul_ai · 2026-10-27
- BELL: Open-Source Stack Combines Browserbase, Eve, iMessage and Stripe Link for Personal Agents — jeff_weinstein · 2026-10-27
- Legora runs 2,000 parallel agent evals on Modal Sandboxes for legal AI — blelbach · 2026-10-27
- Two engineers rebuilt a FedEx supplier's billing core in 3.5 months vs an estimated 7-8 — alex_verem · 2026-10-27
- GPT-6.1 Terra scores 48 on AA Coding Agent Index at one-third the cost of GPT-5.6 Terra — ArtificialAnlys · 2026-10-27
- Shift-1 router cuts coding agent spend up to 55% while keeping SWE-bench accuracy — ypatil125 · 2026-10-27