Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1920 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Smoke Segmentation | MSSDataset (Total) | 18 | Feb 26, 2026 | ||
| Smoke Segmentation | MSSDataset Large | 18 | Feb 26, 2026 | ||
| Visual Number Reasoning | VisNum | 18 | May 26, 2026 | ||
| Pairwise Preference | GenAI Bench (test) | 18 | Jun 15, 2026 | ||
| Pairwise Preference | HPD v3 (test) | 18 | Jun 15, 2026 | ||
| Adversarial Bias Mitigation | WIKI-BALANCE | 18 | Feb 26, 2026 | ||
| Fingerprint Similarity | Qwen 7B 2.5 | 18 | Feb 26, 2026 | ||
| LLM Inference | Transformers (PyTorch) workflow Qwen3 family (inference) | 18 | Feb 26, 2026 | ||
| Offline Reinforcement Learning | Kitchen v0 (partial) | 18 | Feb 26, 2026 | ||
| Offline Reinforcement Learning | Maze2D umaze v1 | 18 | Feb 26, 2026 | ||
| Text Classification | Symptom Diagnosis | 18 | Feb 26, 2026 | ||
| Text Classification | Therapy Diagnosis | 18 | Feb 26, 2026 | ||
| Communication Overhead Measurement | Private Vector-wise Filtering Online Phase | 18 | Feb 26, 2026 | ||
| Attack Success Rate | 20,000 harmful requests and 20,000 jailbreak prompts (test) | 18 | Feb 26, 2026 | ||
| Code Retrieval | CSN-Python Target Queries (test) | 18 | Feb 26, 2026 | ||
| Attack Success Rate | CTCC fingerprinting scenario b | 18 | Feb 26, 2026 | ||
| Code Vulnerability Detection | Devign (test) | 18 | Feb 26, 2026 | ||
| Federated Learning Defense | N-BaIoT Label-Flipping Attack Scenario 1 (test) | 18 | Feb 26, 2026 | ||
| Private Inference | Switch Base | 18 | Feb 26, 2026 | ||
| Private Inference | MoE-Small | 18 | Feb 26, 2026 | ||
| Vulnerability Detection | Dataset (test) | 18 | Feb 26, 2026 | ||
| Function Calling | BFCL | 18 | Feb 26, 2026 | ||
| Tool Use | ToolBench | 18 | Feb 26, 2026 | ||
| Throughput Efficiency | BFCL | 18 | Feb 26, 2026 | ||
| Throughput Efficiency | ToolBench | 18 | Feb 26, 2026 |