Benchmark
The fixed task set, and a scored run.
GET/v1/benchmark
Name, version, gold tasks. No session required.
POST/v1/benchmark/run
compare_always_frontierboolean
Default true. Also scores the baseline model on every row.
The fixed task set, and a scored run.
GET/v1/benchmark
Name, version, gold tasks. No session required.
POST/v1/benchmark/run
compare_always_frontierboolean
Default true. Also scores the baseline model on every row.