The Python client for TensorZero, an open-source platform for optimizing and evaluating LLM applications. Current version 2026.5.0, requires Python >=3.10. Released approximately monthly.
pip install tensorzeroNo compatibility data collected yet for this library.
Verified import paths — ran on the pinned version, not inferred.
Initialize TensorZero client and make a simple inference.
Ensure your gateway and UI authentication settings are consistent.
To restore synchronous writes, set `observability.async_writes = false` in gateway config.
Use `**` for recursive globbing, e.g., `--config-file **/*.toml`.
If you rely on caching, set `cache_options.enabled = 'write_only'` explicitly.
Nest evaluators under functions: `[functions.my_func.evaluators.exact_match]`.
Use `model_name` and `provider_name` instead.
Run `pip install tensorzero` in the correct Python environment (Python >=3.10).
Use `from tensorzero import TensorZeroGateway` instead.
Ensure the config file exists and use `--config-file` with correct path (use `**/*.toml` for recursive search).
Check the response status code and error message. Verify your input schema and ensure provider API keys are set.