LangChain: fine-tuning on agent traces lifts OpenSWE Review precision to 81.5% with 29% fewer tool calls
LangChain · x · 2026-10-01
LangChain ran LangSmith Fine-Tuning on two of its own agents:
- OpenSWE Review precision rose from 62.9% to 81.5% with 29% fewer tool calls per review
- Engine, fine-tuned on Kimi K3, beat both the base model and GPT-5.6 Sol on their issue-detection benchmark
The takeaway: fine-tuning specialized models on your own agent trajectories can outperform general frontier models at lower cost and latency.
Related event: LangChain Launches LangSmith Fine-Tuning to Turn Agent Traces into Models(3 posts)→
More from coding & agent
- Generic API-catalog MCP server vs one tool per job: a context-bloat experiment — MandeepKaur9407 · 2026-11-16
- Berkeley paper: LLMs answer with stale values even when the update is in context — ombre-ai · 2026-11-16
- Effective HTML: 4.2k-star agent skills turn wireframes into self-contained interactive prototypes — tom_doerr · 2026-11-16
- skills-gate: one gate skill fronts 148 agent skills at just 389 tokens — Distinct-Pie2389 · 2026-11-16
- Open-weight Karpo-Decision-31B tops Decision Index at 73.9, one forward pass per answer — CodeByPoonam · 2026-11-16
- shadcn boosts @uisdk: one-shot UI rebuild that's instantly E2E-testable by AI — shadcn · 2026-11-16