LangChain Launches smithtune to Turn Agent Traces Into Fine-Tuned Models
LangChain · x · 2026-10-01
LangChain released smithtune, handling the full lifecycle from agent traces to an SFT'd model on the training/inference provider of your choice, with no manual stitching of steps. The team says data curation is the most important post-training phase, and reports good results using model councils and a CLI-guided coding agent as training copilots. The goal: let builders own their own intelligence by fine-tuning on their own agent data.
Related event: LangChain Launches LangSmith Fine-Tuning to Turn Agent Traces into Models(3 posts)→
More from coding & agent
- dots turns AI into an always-on cloud teammate with its own computer, browser and memory — nickbaumann_ · 2026-10-02
- We shipped 2,000 MCP integrations. Enterprise customers wanted only their own 20 — alexcovo_eth · 2026-10-02
- "The repos are now the prompts, 100%": Ofir Press on AI coding's shifting context — OfirPress · 2026-10-02
- OpenHands announces SOC 2 Type II compliance for enterprise agent security — rajistics · 2026-10-02
- Grady Booch: Coding agents write code well, but architecture still needs human judgment — Grady_Booch · 2026-10-02
- Jev pitches 'decision primitives': models plugging into logic without text — hardimanjames · 2026-10-02