Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 735 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Offline Reinforcement Learning | Maze2D large | 33 | Feb 26, 2026 | ||
| Jailbreak Attack | HarmfulQA | 33 | Feb 26, 2026 | ||
| Over-refusal rate analysis | OR-Bench | 33 | Apr 21, 2026 | ||
| Attribute Inference | Synthetic Dataset | 33 | Feb 26, 2026 | ||
| Text Classification | Yahoo | 33 | Mar 24, 2026 | ||
| Hallucination detection | Dolly AC (test) | 33 | Feb 26, 2026 | ||
| Multi-hop QA Retrieval | 2WikiMultihopQA (test) | 33 | May 22, 2026 | ||
| Text Generation | Text Generation | 33 | Feb 26, 2026 | ||
| Competitive Programming | LiveCodeBench Pro 25Q2 | 33 | Mar 20, 2026 | ||
| Competitive Programming | LiveCodeBench Pro 25Q1 | 33 | Mar 20, 2026 | ||
| General Utility Evaluation | MT_Bench | 33 | Jun 1, 2026 | ||
| Planning | PDDL | 33 | Jun 4, 2026 | ||
| Multilingual Text Embedding | MTEB Multilingual | 33 | Jun 25, 2026 | ||
| Personalized Dialogue Response Generation | PersonaMem 1.0 | 33 | Feb 26, 2026 | ||
| Tabular Question Answering | ReasonTabQA 1.0 (Overall) | 33 | Feb 26, 2026 | ||
| Sequential Knowledge Editing | ZsRE sequential editing 10,000 Samples | 33 | Feb 26, 2026 | ||
| Sequential Knowledge Editing | CounterFact sequential editing 10,000 Samples | 33 | Feb 26, 2026 | ||
| Membership Inference Attack | WikiMIA-25 | 33 | Feb 26, 2026 | ||
| Long-context Question Answering | LongBench V2 | 33 | May 8, 2026 | ||
| Code Generation | MBPP+ | 33 | Jun 9, 2026 | ||
| Single-hop Question Answering | Natural Questions (NQ) (test) | 33 | Mar 5, 2026 | ||
| Machine Translation | Tatoeba it->en | 33 | Feb 26, 2026 | ||
| Machine Translation | Tatoeba en->it | 33 | Feb 26, 2026 | ||
| Safety Dialogue Evaluation | SafeDialBench | 33 | Feb 26, 2026 | ||
| Mathematical Reasoning | GSM8K Var | 33 | Feb 26, 2026 |