Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1490 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Script confusion mitigation | FLEURS sr-cyrl (test) | 21 | Feb 26, 2026 | ||
| Script confusion mitigation | FLEURS sr-latn (test) | 21 | Feb 26, 2026 | ||
| Factuality Evaluation | Held-out Dataset | 21 | Feb 26, 2026 | ||
| Language Understanding | MMLU en | 21 | Mar 6, 2026 | ||
| Mathematical Reasoning | GSM8K | 21 | Feb 26, 2026 | ||
| Mathematical Reasoning | MATH500 | 21 | Feb 26, 2026 | ||
| Question Answering | TYDIQAGP | 21 | Feb 26, 2026 | ||
| Question Answering | COQA | 21 | Feb 26, 2026 | ||
| Question Answering | TriviaQA | 21 | Feb 26, 2026 | ||
| Question Answering | TRUTHFULQA | 21 | Feb 26, 2026 | ||
| Question Answering | HotpotQA | 21 | Feb 26, 2026 | ||
| Pre-training Data Detection | WikiMIA Original Length 32 | 21 | Feb 26, 2026 | ||
| Mathematical Reasoning | AMC | 21 | Feb 26, 2026 | ||
| Mathematical Reasoning | AIME 2025 | 21 | Feb 26, 2026 | ||
| Mathematical Reasoning | AIME 2024 | 21 | Feb 26, 2026 | ||
| Knowledge Evaluation | KMMLU, KMMLU Redux, KMMLU Pro, CLIcK, KoBALT, MMLU Pro, GPQA Diamond | 21 | Feb 26, 2026 | ||
| SQL execution performance | BIRD n=1534 | 21 | Feb 26, 2026 | ||
| Mathematical Reasoning | SVAMP | 21 | Feb 26, 2026 | ||
| Multi-hop Question Answering | MusiQue answerable setting | 21 | Feb 26, 2026 | ||
| Multi-hop Question Answering | HotpotQA distractor setting | 21 | Feb 26, 2026 | ||
| Mathematical Reasoning | MATH-500 (train) | 21 | Feb 26, 2026 | ||
| Video Deepfake Detection | Celeb-DF (CDF) | 21 | Feb 26, 2026 | ||
| Question Answering | en multifield | 21 | Feb 27, 2026 | ||
| Summarization | news multi | 21 | Feb 27, 2026 | ||
| LLM Inference | Aggregate Mean over Alpaca, CodeAlpaca, HumanEval, LiveCodeBench, Math500, MBPP, MT-Bench | 21 | Feb 26, 2026 |