Skip to content

Leaderboard Entries Models Changelog Docs

Sign in Get API key

Compare

Both entries need a published score on claude-fable-5 before they can be compared.

evalnessMeasured, not claimed.

Every entry is scored against a named model version, with confidence intervals and the run that produced it.

Registry

Leaderboard All entries Compare entries Tracked models Model changelog

Build

API and MCP docs Submit an entry API keys REST endpoint

About

Methodology Privacy Terms Sign in