From the source
LangChain announced LangSmith Fine-Tuning, now in public beta, which uses an open-source CLI called smithtune to read LangSmith traces, convert successful runs into a supervised fine-tuning dataset, and train on Baseten Loops.
The CLI can deploy the resulting checkpoint to a Baseten Dedicated Inference deployment.
Training runs on dedicated GPUs in the user's own Baseten workspace, and billing goes to the user's Baseten account.
