MMLU
Harbor adapterMMLU evaluates English language models on knowledge and reasoning across 57 diverse subjects using a multiple-choice format with over 14k samples.
- Samples
- 14,042
- Released
- 2020
- Scores
- 209
Available in EleutherAI lm-evaluation-harness as task family mmlu-redux (57 tasks).
$ npm i -g lettokenburn-cli$ ltb bench pull mmlu-redux --n 50$ ltb bench run ./mmlu-redux -a claude-code -m anthropic/claude-opus-4-8$ ltb bench push
Runs happen on your machine with your own subscription; only the score summary is pushed. CLI docs
No sourced scores recorded yet.
Every number links to where it was published. Low-confidence rows are muted and need per-cell verification.
Available in EleutherAI lm-evaluation-harness as task family mmlu-redux (57 tasks).