Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 761 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Tool Learning under Instructions Beyond Tool Capabilities | NoisyToolBench IBTC 1.0 (test) | 32 | Apr 30, 2026 | ||
| Tool Learning under Instruction with Error | NoisyToolBench IwE 1.0 (test) | 32 | Apr 30, 2026 | ||
| Tool Learning under Instruction with Multiple Requests | NoisyToolBench IMR 1.0 (test) | 32 | Apr 30, 2026 | ||
| Tool Learning under Instruction with Missing Key Information | NoisyToolBench IMKI 1.0 (test) | 32 | Apr 30, 2026 | ||
| Tool-using | NoisyToolBench IBTC | 32 | Apr 30, 2026 | ||
| Tool-using | NoisyToolBench IwE | 32 | Apr 30, 2026 | ||
| Tool-using | NoisyToolBench IMR | 32 | Apr 30, 2026 | ||
| Tool-using | NoisyToolBench IMKI | 32 | Apr 30, 2026 | ||
| Software Engineering | SWE-bench verified (All) | 32 | Apr 29, 2026 | ||
| Paper Reproduction | RE-Bench Weighted | 32 | Apr 28, 2026 | ||
| Paper Reproduction | RE-Bench Hard tier | 32 | Apr 28, 2026 | ||
| Paper Reproduction | RE-Bench Medium tier | 32 | Apr 28, 2026 | ||
| Paper Reproduction | RE-Bench Easy tier | 32 | Apr 28, 2026 | ||
| Jailbreak Detection | Jailbreak data (70/30 stratified) | 32 | Apr 28, 2026 | ||
| Dialogue Generation | UltraChat | 32 | Apr 28, 2026 | ||
| Mathematical Reasoning | GSM8K | 32 | Apr 28, 2026 | ||
| Question Answering | HotpotQA (test) | 32 | May 1, 2026 | ||
| Language Modeling | GPT-2 | 32 | Apr 28, 2026 | ||
| Probabilistic Electricity Price Forecasting | German electricity price day-ahead market n = 736 (First sub-sample: 2018-12-27 to 2020-12-31) | 32 | Apr 27, 2026 | ||
| Prediction Region Estimation | Synthetic data 100 seeds (test) | 32 | Apr 24, 2026 | ||
| Prevalence Estimation | ACS (American Community Survey) Out-of-Distribution | 32 | Apr 24, 2026 | ||
| Prevalence Estimation | ACS (American Community Survey) In-Distribution | 32 | Apr 24, 2026 | ||
| Out-of-Distribution Detection | ImageNet1K-OpenOOD ImageNet-O | 32 | Jun 17, 2026 | ||
| Visual Question Answering | ChartQA | 32 | Jun 19, 2026 | ||
| GUI Navigation | OSWorld Verified | 32 | Apr 24, 2026 |