LFM2.5-24B LoRA adapters for Pi and Claude Code cut tool calls up to 40%
yb2698 · x · 2026-10-30
The harness-hub team released LoRA adapters trained on top of LiquidAI's LFM2.5-24B, targeting two open-source agent harnesses — Pi and Claude Code — for more efficient agentic development, with up to 40% fewer tool calls on Pi.
Training recipe: following a public recipe, they split a single B300 GPU in half — one half serving vLLM, the other running training — finishing the whole run in 24 hours on one GPU.
Related event: LFM2.5-24B gets LoRA adapters for Pi and Claude Code(4 posts)→
More from coding & agent
- A 4B Model Verifies a Coding Agent 70x Its Size at Just 4% of Steps, 9-13% Extra Cost — petrusenko_max · 2026-10-30
- Static RAG eval sets go stale: treat your eval dataset as a pipeline, not a file — paulabartabajo_ · 2026-10-30
- What took a week now takes ~5 minutes with Sol or Opus coding agents — D3VAUX · 2026-10-30
- dhh: 'Review every line' is AI psychosis — the future is verifying outcomes, not code — CtrlAltDwayne · 2026-10-30
- Do you need an AI QA agent writing Playwright tests when AI already reviews PRs? — Infinite-Fly-2756 · 2026-10-30
- Octoverse: Claude Code tops fastest-growing repos; helium lands at #9 — uwukko · 2026-10-30