Loading the SOTA2 catalog…
SOTA2 Research · benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
| Task | Dataset | Domain | Trend | Results | Last Update |
|---|---|---|---|---|---|
| Text-to-Image Alignment | MS COCO | 68 | Jul 9, 2026 | ||
| Instruction Following | IFBench | 68 | Jun 16, 2026 | ||
| Human Pose Estimation | MM-Fi | 68 | May 12, 2026 | ||
| Knowledge-grounded reasoning | WISE |
| 68 |
| Mar 31, 2026 |
| GUI Grounding | UI-Vision | 68 | May 27, 2026 |
|---|
| Artifact Detection | OpenMMSec | 68 | Feb 26, 2026 |
|---|
| Text-to-Image Generation | DPG-Bench (test) | 68 | May 18, 2026 |
|---|
| Text-to-SQL | BIRD | 68 | Jun 1, 2026 |
|---|
| Multi-Domain Knowledge | MMLU | 68 | Jun 12, 2026 |
|---|
| Classification | FMNIST | 68 | Jul 7, 2026 |
|---|
| Hallucination Detection | Company | 68 | Feb 26, 2026 |
|---|
| Instruction Following | CF-Bench | 68 | May 11, 2026 |
|---|
| Clustering | CLSClusteringS2S | 68 | Feb 26, 2026 |
|---|
| Model Calibration | CIFAR10 (test) | 68 | May 26, 2026 |
|---|
| Face Forgery Detection | Protocol 2 Forgery Type Incremental: Hybrid, FR, FS, EFS | 68 | Apr 20, 2026 |
|---|
| Face Forgery Detection | Protocol 1 Dataset Incremental: SDv21, FF++, DFDCP, CDF | 68 | Apr 20, 2026 |
|---|
| Novel View Synthesis | Mip-NeRF 360 24-view | 68 | Jul 2, 2026 |
|---|
| Code Generation | LiveCodeBench | 68 | Feb 27, 2026 |
|---|
| Image Quality Assessment | AGIQA-1K | 68 | Jul 8, 2026 |
|---|
| Multi-Hop Knowledge Graph Question Answering | GrailQA | 68 | Jun 2, 2026 |
|---|
| Mathematical Reasoning | AMC23 (test) | 68 | Jun 16, 2026 |
|---|
| Scientific Reasoning | GPQA Diamond | 68 | Apr 10, 2026 |
|---|
| Locomotion | D4RL halfcheetah-medium-replay | 68 | May 19, 2026 |
|---|
| Image Classification | Downstream Datasets Average | 68 | Jun 2, 2026 |
|---|
| Multi-turn Conversation Evaluation | MT-bench | 68 | May 12, 2026 |
|---|