Recipe — port a LangSmith eval dataset and judge into Stratix.
Export LangSmith dataset to JSONL.
Upload to Stratix as private benchmark.
Re-author judge using Stratix judge builder.
(Recommended) GEPA-optimize against your LangSmith labels.
client.benchmarks.upload("./langsmith-export.jsonl", name="ported-from-langsmith")
Migration: From LangSmith
Last updated 1 month ago
Was this helpful?