How to go from your agent's traces to a fine-tuned model in one workflow
LangChain launches LangSmith fine-tuning in public beta with SmithTune, a CLI to post-train models from agent traces.
“today we're launching LangSmith fine-tuning in public beta with SmithTune, a CLI to allow you to post-train models from your LangSmith traces in one workflow”
LangChain released SmithTune, a CLI that turns LangSmith agent traces into supervised fine-tuning datasets and runs the full workflow from trace selection through model-council triage, training on providers like Base10 or Fireworks, and evaluation. It matters because it lowers the barrier to distilling and specializing models on real agent trajectories, tightening the loop between observability and post-training.