"Inference Engineering" is now available. Get your copy here
changelog / post

Per-request log filtering

Go back

Every predict call now returns a unique request ID in the X-Baseten-Request-Id response header. Use this ID to filter your model's logs to a single request, cutting through the noise when debugging individual predictions in production. Works across HTTP, gRPC, and async predict calls. Requires Truss 0.15.5 or later.

For more information, see Logs.