v2
latestOpenAPI 3.1.02026-08-05267431678.1 KBfelix
Upload Evaluation Results
Upload manually computed evaluation results so they appear in the UI.
post/felix/evaluations/upload
Request body
model_idstring required
Model identifier used for the evaluation
dataset_namestring required
Pioneer dataset name the evaluation was run against
dataset_versionstring nullable
Dataset version (latest if omitted)
project_idstring nullable
Optional project ID to associate the evaluation with
model_namestring nullable
Optional human-readable model name for the UI
status'complete' | 'failed' | 'errored'
Final evaluation status. 'errored' is normalized to 'failed'.
f1_scorenumber nullable
precision_scorenumber nullable
recall_scorenumber nullable
accuracynumber nullable
validation_lossnumber nullable
subset_accuracynumber nullable
hamming_lossnumber nullable
bleu_scorenumber nullable
rouge_l_scorenumber nullable
total_tokensinteger nullable
total_cost_usdnumber nullable
total_latency_msinteger nullable
max_examplesinteger nullable
max_tokensinteger nullable
seedinteger nullable
configobject nullable
Optional evaluation metadata stored in model_evaluations.config, for example threshold sweeps, schemas, judge criteria, or system prompts.
error_messagestring nullable
error_countinteger nullable
error_samplestring nullable
failed_examplesobject[] nullable
sample_countinteger nullable
evaluation_time_msinteger nullable
completed_atstring date-time nullable
predictionsobject[] nullable
Response
Successful Response
idstring required
user_idstring required
project_idstring nullable
model_idstring required
dataset_namestring required
dataset_versionstring required
model_namestring nullable
f1_scorenumber nullable
precision_scorenumber nullable
recall_scorenumber nullable
accuracynumber nullable
validation_lossnumber nullable
subset_accuracynumber nullable
hamming_lossnumber nullable
bleu_scorenumber nullable
rouge_l_scorenumber nullable
total_tokensinteger nullable
total_cost_usdnumber nullable
total_latency_msinteger nullable
max_examplesinteger nullable
max_tokensinteger nullable
seedinteger nullable
configobject nullable
statusstring required
job_referencestring nullable
error_messagestring nullable
error_countinteger nullable
error_samplestring nullable
sample_countinteger nullable
evaluation_time_msinteger nullable
created_atstring date-time required
completed_atstring date-time nullable
failed_examplesobject[] nullable
predictionsobject[] nullable