Skip to main content
Unlisted page
This page is unlisted. Search engines will not index it, and only users having a direct link can access it.

Create model evaluation

POST 
/models/evaluations

Starts a new benchmark evaluation run for a deployed fine-tuned model. Evaluation scores are returned per benchmark and reflect model performance on each task. Available evaluation types include mix_eval and task-specific benchmarks. Evaluations are available for fine-tuned models only; imported LoRA adapters and base models are not supported.

Request​

Responses​

OK