LangChain launches LangSmith Fine-Tuning and smithtune CLI to turn traces into tuned models
LangChain · x · 2026-09-25
LangChain introduced LangSmith Fine-Tuning and the smithtune CLI (public beta), a single workflow for fine-tuning:
- Identify and export trajectories from LangSmith
- Transform traces into SFT-ready training examples
- Fine-tune open models via Baseten and Fireworks AI
- Evaluate back in LangSmith
The pitch: use production traces to train specialized models that cut cost and latency.
More from coding & agent
- Four steps to stop Opus 5.5 from wasting tokens: effort, prompt audit, CLAUDE.md rules — dr_cintas · 2026-09-25
- Nat Friedman fires back at Verge: those 'leaked secrets' are just files on the user's VM, intended behavior — dtrinh · 2026-09-25
- ServingStudio: simulate LLM serving configs before burning expensive GPU time — bariskasikci · 2026-09-25
- issue-graph prioritizes your backlog to coordinate coding agents, now on GitHub — Chamoen · 2026-09-25
- Claude now spots bugs, gathers logs, and asks users to submit packaged reports — ColleenMBrady · 2026-09-25
- SkillRL (NeurIPS 2026): 7B model beats GPT-4o by 41% via recursive skill evolution — cihangxie · 2026-09-25