Skip to main content
RECOGNITION · PROGRAMMES · ECOSYSTEM · TRUST FinanceGPT Labs
FinanceGPT Developer Platform

Model evaluations

Compare public sealed evaluation cards for governed published models. Unlike tasks and unlike evaluation suites are not ranked as though they were directly comparable.

Public evaluations
0
Published + hash-matched
Raw inputs exposed
0
Never
Execution manifests
0
Withheld
Auto promotion
0
Evaluation is evidence only

Published evaluation cards

Metrics are grouped by each model evaluation suite. HF3 does not rank unlike tasks or unlike suites as though they were directly comparable.
ModelTaskSuiteResultEvaluatedAction
No public published model currently has a matching sealed evaluation card. HF3 will surface one automatically after an explicitly published model carries valid evaluation evidence.