Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1247 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Personalization Editing | UPQA balanced 100-sample | 24 | Feb 26, 2026 | ||
| General Capability Evaluation (Held-out Benchmarks) | OlmoBaseEval LBPP BBH MMLU Pro MC Deepmind Math (HeldOut) | 24 | Feb 26, 2026 | ||
| Jailbreak Attack | Claude 3.5 | 24 | May 19, 2026 | ||
| Mathematical Reasoning | GSM-Symbolic (test) | 24 | Feb 26, 2026 | ||
| Speech Recognition | LibriSpeech 960 clean (test) | 24 | May 29, 2026 | ||
| Text-to-Speech | InstructTTSEval ZH | 24 | Apr 10, 2026 | ||
| Question Answering | Merged QA HotpotQA, NarrativeQA, WikiHop (test) | 24 | Feb 26, 2026 | ||
| Question Answering | HotpotQA (test) | 24 | Feb 26, 2026 | ||
| Mathematical Reasoning | GSM8K (test) | 24 | Feb 26, 2026 | ||
| Deep Research | DeepResearch Bench official 100-task-subset 1.0 | 24 | Feb 26, 2026 | ||
| Financial Analysis | MCP-Universe Financial Analysis | 24 | Feb 26, 2026 | ||
| Repository Management | MCP-Universe Repository Management | 24 | Feb 26, 2026 | ||
| Location Navigation | MCP-Universe Location Navigation | 24 | Feb 26, 2026 | ||
| Tool Reasoning | ToolBench (G3) | 24 | Feb 26, 2026 | ||
| Tool Reasoning | ToolBench G2 | 24 | Feb 26, 2026 | ||
| Tool Reasoning | ToolBench (G1) | 24 | Feb 26, 2026 | ||
| Deep Search | GAIA text-only (val) | 24 | Feb 26, 2026 | ||
| Mathematical Reasoning | MATH (test) | 24 | Feb 26, 2026 | ||
| Mathematical Reasoning | MATH (test) | 24 | Feb 26, 2026 | ||
| Mathematical Reasoning | GSM8K zero-shot (test) | 24 | Feb 26, 2026 | ||
| Dialogue Memory Accuracy | LongMemEval-S (N=500) | 24 | Apr 21, 2026 | ||
| Question Answering | StrQA | 24 | Feb 26, 2026 | ||
| Multi-Domain Knowledge | GPQA | 24 | Feb 26, 2026 | ||
| Multi-Domain Knowledge | MMLU-Pro | 24 | Feb 26, 2026 | ||
| Math & Logic | MUSR | 24 | Feb 26, 2026 |