OpenForgeRL: Train Harness-native Agents in Any Environment
Modern AI agents rely on elaborate inference harnesses such as Claude Code, Codex, and OpenClaw to drive multi-turn reasoning, tool use, and access to external systems. While powerful, these complex harnesses also make agents hard to train end-to-end with open infrastructure, whose SFT/RL stacks cannot natively express stateful, multi-process harness inference. To address this, we present OpenForgeRL, an open-source framework for training harness-based agents end-to-end in diverse environments. OpenForgeRL achieves this with a lightweight proxy that serves the harness's model calls while recor
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.
- FuzzySimilar title/name (fuzzy) · 87%CodeForge-15B →
“Fuzzy title match (0.94): “OpenForgeRL: Train Harness-native Agents in Any Environment” ≈ “CodeForge-15B””
- PossiblePossibly related (embedding) · 25%OpenHands/OpenHands →
“Possibly related via embedding similarity 0.55 (not asserted). Timestamp check: artifact slightly before paper (-21d).”
- FuzzySimilar title/name (fuzzy) · 87%NirDiamant/GenAI_Agents →
“Fuzzy title match (0.94): “OpenForgeRL: Train Harness-native Agents in Any Environment” ≈ “NirDiamant/GenAI_Agents””
- FuzzySimilar title/name (fuzzy) · 87%strands-agents/harness-sdk →
“Fuzzy title match (0.94): “OpenForgeRL: Train Harness-native Agents in Any Environment” ≈ “strands-agents/harness-sdk””
- FuzzySimilar title/name (fuzzy) · 84%Unity-Technologies/ml-agents →
“Fuzzy title match (0.92): “OpenForgeRL: Train Harness-native Agents in Any Environment” ≈ “Unity-Technologies/ml-agents””
- FuzzyOverlapping authors or contributors · 62%pytorch/pytorch →
“Shared author/contributor keys: zou”
- FuzzyOverlapping authors or contributors · 62%DietrichGebert/ponytail →
“Shared author/contributor keys: cheng”
- LinkedLinked via arxiv author · 85%Xiao Yu →
“OpenForgeRL: Train Harness-native Agents in Any Environment”