RL-Trained LoRA Cuts LFM2.5-24B Tool Calls by 39% in Agent Harnesses
amaarora · x · 2026-10-30
Liquid AI released LoRA adapters for LFM2.5-24B-A2B, RL-trained to run more efficiently inside open-source agent harnesses: same task-solving accuracy with 39-40% fewer tool calls in the pi harness.
- Adapters available for both pi and Claude Code on Hugging Face (harness-hub collection), with GGUF quantized versions (18.7MB)
- Recipe: Liquid AI base + badlogic's pi + TRL AsyncGRPO, trained in 24h on a single GPU
- The idea: use RL to specialize the model's calling patterns for a specific harness, trimming redundant steps without sacrificing accuracy
Related event: LFM2.5-24B Harness Adapters Cut Tool Calls ~39% via LoRA+RL(6 posts)→
More from coding & agent
- Data leakage hit 648 papers across 30 fields; autoarxiv now auto-reproduces arXiv papers — pandeyparul · 2026-10-30
- METR trial: AI made experienced devs 19% slower while they felt 20% faster — srimisra · 2026-10-30
- Clairvoyance: free desktop tool wires Claude Code, Codex and other agents into your daily workflow — draginol · 2026-10-30
- Same Prompt, Same Data: Claude's Fabric App Beats Its Dashboard at One-Fifth the Cost — adnan_hashmi · 2026-10-30
- Antigravity's Google Slides Plugin Audits an 80-Slide Deck Against Docs — rseroter · 2026-10-30
- OpenAI hosts Cloud Native Hackathon Nov 9; hack turns M5Stack into a k8s desk pager — OpenAIDevs · 2026-10-30