Skip to content
GitHub
View on GitHub

EvalRowResult

One model interaction: the prompt, the raw response, its parsed structure (thinking / answer / tool calls), a score, and free-form metadata.

from modal_training_gym.common.eval import EvalRowResult

One model interaction: the prompt, the raw response, its parsed structure (thinking / answer / tool calls), a score, and free-form metadata.

EvalRowResult(**data)
ParameterTypeDefaultDescription
AttributeTypeDefaultDescription
scorefloat
promptstr
responsestr
parsed_responseParsedResponse | None
metadatadict[str, Any]

Source: modal_training_gym/common/eval.py