Authorizations
Body
application/json
Request model for binary evaluation of LLM responses against specified criteria.
A list of chat messages
Minimum length:
2
Criteria used for evaluation. Begins with one of the following: Response passes if, Response fails if, Tool call passes if, Tool call fails if, Agent passes if, Agent fails if
System message is separate for Anthropic-style LLM calls. Optional.
List of tools available for the assistant to call. Optional.
The model core for reward evaluation. Defaults to align-20250503 if not specified.
Available options:
align-20250529
, align-lightning-20250731