LLM Stats leaderboard
34 scores · 34 models · first recorded Oct 2, 2026
higher is better · scale 0–1 · record SuperGPQA
Best reported score
73.6%Qwen3.7 Max · LLM Stats leaderboard
37 models tested · 2 sources · 5 source documents · latest report Jun 24, 2026
Scores stay separate per source — never merged into one cross-source ranking.
34 scores · 34 models · first recorded Oct 2, 2026
higher is better · scale 0–1 · record SuperGPQA
| Model | Value (%) | Reported | Score type | Recorded |
|---|---|---|---|---|
| Seed 2.1 Pro | 70.8% | Jun 24, 2026 | unspecified | Oct 2, 2026 |
| Seed 2.1 Turbo | 67.4% | Jun 24, 2026 | unspecified | Oct 2, 2026 |
| Qwen3.7-Plus | 71.4% | May 31, 2026 | unspecified | Oct 2, 2026 |
| Qwen3.7 Max | 73.6% | May 19, 2026 | unspecified | Oct 2, 2026 |
| Qwen3.6-27B | 66% | Apr 21, 2026 | unspecified | Oct 2, 2026 |
| Qwen3.6-35B-A3B | 64.7% | Apr 16, 2026 | unspecified | Oct 2, 2026 |
| Qwen3.6 Plus | 71.6% | Apr 2, 2026 | unspecified | Oct 2, 2026 |
| Qwen3.5-9B | 58.2% | Mar 2, 2026 | unspecified | Oct 2, 2026 |
| Qwen3.5-4B | 52.9% | Mar 2, 2026 | unspecified | Oct 2, 2026 |
| Qwen3.5-2B | 37.5% | Mar 2, 2026 | unspecified | Oct 2, 2026 |
| Qwen3.5-0.8B | 21.3% | Mar 2, 2026 | unspecified | Oct 2, 2026 |
| Qwen3.5-122B-A10B | 67.1% | Feb 24, 2026 | unspecified | Oct 2, 2026 |
| Qwen3.5-27B | 65.6% | Feb 24, 2026 | unspecified | Oct 2, 2026 |
| Qwen3.5-35B-A3B | 63.4% | Feb 24, 2026 | unspecified | Oct 2, 2026 |
| Qwen3.5-397B-A17B | 70.4% | Feb 16, 2026 | unspecified | Oct 2, 2026 |
| Qwen3 Max | 65.1% | Dec 15, 2025 | unspecified | Oct 2, 2026 |
| Qwen3 VL 235B A22B Thinking | 64.3% | Sep 22, 2025 | unspecified | Oct 2, 2026 |
| Qwen3 VL 235B A22B Instruct | 60.4% | Sep 22, 2025 | unspecified | Oct 2, 2026 |
| Qwen3 VL 32B Thinking | 59% | Sep 22, 2025 | unspecified | Oct 2, 2026 |
| Qwen3 VL 30B A3B Thinking | 56.4% | Sep 22, 2025 | unspecified | Oct 2, 2026 |
| Qwen3 VL 32B Instruct | 54.6% | Sep 22, 2025 | unspecified | Oct 2, 2026 |
| Qwen3 VL 30B A3B Instruct | 53.1% | Sep 22, 2025 | unspecified | Oct 2, 2026 |
| Qwen3 VL 8B Thinking | 51.2% | Sep 22, 2025 | unspecified | Oct 2, 2026 |
| Qwen3 VL 4B Thinking | 46.8% | Sep 22, 2025 | unspecified | Oct 2, 2026 |
| Qwen3 VL 8B Instruct | 44.5% | Sep 22, 2025 | unspecified | Oct 2, 2026 |
| Qwen3 VL 4B Instruct | 40.3% | Sep 22, 2025 | unspecified | Oct 2, 2026 |
| Qwen3-Next-80B-A3B-Thinking | 60.8% | Sep 10, 2025 | unspecified | Oct 2, 2026 |
| Qwen3-Next-80B-A3B-Instruct | 58.8% | Sep 10, 2025 | unspecified | Oct 2, 2026 |
| Kimi K2-Instruct-0905 | 57.2% | Sep 5, 2025 | unspecified | Oct 2, 2026 |
| Qwen3-235B-A22B-Thinking-2507 | 64.9% | Jul 25, 2025 | unspecified | Oct 2, 2026 |
| Qwen3-235B-A22B-Instruct-2507 | 62.6% | Jul 22, 2025 | unspecified | Oct 2, 2026 |
| Kimi K2 Instruct | 57.2% | Jul 11, 2025 | unspecified | Oct 2, 2026 |
| Kimi K2 Base | 44.7% | Jul 11, 2025 | unspecified | Oct 2, 2026 |
| Qwen3 235B A22B | 44.1% | Apr 29, 2025 | unspecified | Oct 2, 2026 |
4 scores · 4 models · first recorded Oct 2, 2026
higher is better · accuracy · scale 0–100 · record SuperGPQA
| Model | Value (%) | Subset / variant | Test conditions | Reported by | Reported | Document |
|---|---|---|---|---|---|---|
| Qwen3.5-397B-A17B | 70.4% | super_gpqa | thinking (temp 0.6, top_p 0.95, top_k 20) | Qwen | Feb 16, 2026 | qwen3_5_model_card |
| Seed2.0 Pro | 68.7% | super_gpqa | as printed in Table 3 (Large Models) | ByteDance | Feb 14, 2026 | bytedance_seed2_0_model_card |
| Seed2.0 Lite | 57.9% | super_gpqa | as printed in Table 4 (Efficient Models) | ByteDance | Feb 14, 2026 | bytedance_seed2_0_model_card |
| Seed2.0 Mini | 49.8% | super_gpqa | as printed in Table 4 (Efficient Models) | ByteDance | Feb 14, 2026 | bytedance_seed2_0_model_card |
| Model | Value (%) | Reported | Score type | Recorded |
|---|---|---|---|---|
| Qwen3.5-397B-A17B | 70.4% | Feb 16, 2026 | self-reported | Oct 2, 2026 |
| Seed2.0 Pro | 68.7% | Feb 14, 2026 | self-reported | Oct 2, 2026 |
| Seed2.0 Lite | 57.9% | Feb 14, 2026 | self-reported | Oct 2, 2026 |
| Seed2.0 Mini | 49.8% | Feb 14, 2026 | self-reported | Oct 2, 2026 |
every document that defines or reports scores
| other | – | SuperGPQA– |
| other | ByteDance | Seed2.0 (Pro / Lite / Mini)Feb 14, 2026 |
| other | DeepSeek | DeepSeek-V4-ProApr 22, 2026 |
| other | Qwen | Qwen3.5-397B-A17BFeb 16, 2026 |
| other | – | SuperGPQA– |
Records are located via the tracking corpus (ktwu01/benchmark-radar live dataset). Every row cites its source document.