MMLU-Redux Benchmark

Explore score history ↗

MMLU-Redux is an AI benchmark in the Benchmark Radar catalog.

Released: 2024-06-07
Openness: unknown
Importer: Benchmark Saturation

Source provenance

Original evidence

MMLU-Redux results and reported scores

Reported benchmark scores · Benchmark Saturation

Reported scores: 5

Model	Organization	Reported value	Reported	Evidence
Claude opus 4	—	94.2	—	evidence
Claude sonnet 4	—	93.6	—	evidence
Kimi K2-Thinking-0905	—	94.4	—	evidence
Qwen3 VL 235B A22B Thinking	—	93.7	—	evidence
Qwen3-235B-A22B-Thinking-2507	—	93.8	—	evidence

Scores are partitioned by the source that reported them and are never merged into a single cross-source ranking, because the sources measure different things and say so.

Which sources cite MMLU-Redux?

MMLU-Redux