Multi-harness RL guide: LFM2.5 jumps 42% to 54% with 31% fewer tool calls
SergioPaniego · x · 2026-10-01
Adithya S K released an open guide to multi-harness RL, built on the observation that the same model behaves differently across agent harnesses. The method trains any model with RL on any task set inside real-world harnesses like Claude Code, Codex, and OpenCode, without changing harness or training code. Trained across four harnesses, LFM2.5-2.6B improved from 42% to 54% while making 31% fewer tool calls.
Related event: Multi-Harness RL Boosts LFM2.5 from 42% to 54%(2 posts)→
More from coding & agent
- Early Hands-On: OpenAI's Dots Agent Impresses With Speed and First-Try Accuracy — billyjhowell · 2026-10-02
- Google launches Advent of Agents Season 3: 31 days of free agent-building lessons — Saboo_Shubham_ · 2026-10-02
- MINTEval, accepted at NeurIPS, shows LLM agents fail at tracking evolving contexts — EliasEskin · 2026-10-02
- Claude Code lead on the sassy status dot: it might just be busy with something else — cyrus_zei · 2026-10-02
- LlamaIndex Launches Extract v2.5, Beats Claude and GPT at 30%-4x Lower Cost — llama_index · 2026-10-02
- Dev Finds First CVE in Ghost: CVSS 8.8 Memory Bug, PoC Built by MiniMax M3 — DanielLockyer · 2026-10-02