Other · model
Seed 2.1 Turbo benchmark scores
As of 2026-10-07, Seed 2.1 Turbo (Other) has measured scores on 9 benchmarks and estimated scores on 35 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 85.6% | estimated ± 5.0 pp, medium confidence |
| AA-SciCode | 48.3% | estimated ± 3.0 pp, high confidence |
| Artificial Analysis Intelligence Index | 22.5% | estimated ± 3.9 pp, medium confidence |
| GPQA Diamond | 83.9% | estimated ± 12.9 pp, low confidence |
| HLE | 30.5% | estimated ± 7.8 pp, medium confidence |
| LiveCodeBench v6 | 88.5% | estimated ± 3.8 pp, low confidence |
| AA-Omniscience Accuracy | 25.6% | estimated ± 6.0 pp, medium confidence |
| Terminal-Bench 2.1 | 67.6% | measured |
| AA Coding Index | 62.3% | estimated ± 4.6 pp, high confidence |
| AA-MMMU-Pro | 80.9% | estimated ± 2.2 pp, high confidence |
| SWE-bench Verified | 78.8% | estimated ± 2.4 pp, high confidence |
| SWE-bench Pro | 59.0% | estimated ± 3.6 pp, high confidence |
| MMLU-Pro | 84.5% | estimated ± 6.8 pp, medium confidence |
| GPQA | 86.6% | estimated ± 2.6 pp, high confidence |
| Terminal-Bench 3.0 | 7.3% | estimated ± 9.1 pp, low confidence |
| Vals SWE-bench | 77.5% | estimated ± 5.4 pp, medium confidence |
| Terminal-Bench 2.1 (Vals AI) | 58.2% | estimated ± 7.4 pp, medium confidence |
| AA Terminal-Bench 2.1 | 67.6% | estimated ± 3.5 pp, high confidence |
| AA Terminal-Bench 4.0 | 4.6% | estimated ± 8.9 pp, medium confidence |
| DeepSWE | 41.3% | estimated ± 9.6 pp, medium confidence |
| CharXiv | 82.5% | measured |
| CharXiv w/o tools | 83.7% | estimated ± 1.7 pp, medium confidence |
| C-Eval | 91.5% | estimated ± 0.7 pp, medium confidence |
| MathVision | 90.1% | measured |
| MMLU-Redux | 94.1% | estimated ± 0.9 pp, high confidence |
| MMMU-Pro | 80.1% | measured |
| NL2Repo | 43.7% | measured |
| ScreenSpot Pro | 74.7% | estimated ± 4.5 pp, medium confidence |
| SuperGPQA | 67.4% | measured |
| V* | 96.8% | estimated ± 5.1 pp, low confidence |
| VideoMMMU | 86.6% | estimated ± 1.0 pp, high confidence |
| ERQA | 71.3% | measured |
| MedXpertQA (MM) | 74.8% | estimated ± 2.4 pp, high confidence |
| MMMU | 85.7% | estimated ± 1.1 pp, low confidence |
| OpenHarmony Bench | 52.0% | estimated ± 2.7 pp, high confidence |
| BabyVision | 62.9% | measured |
| MMVU | 77.8% | estimated ± 2.9 pp, medium confidence |
| SimpleVQA | 74.4% | estimated ± 2.5 pp, low confidence |
| ZeroBench | 11.0% | measured |
| MMMU-Pro w/ Python | 81.4% | estimated ± 0.6 pp, high confidence |
| MathVision w/ Python | 95.2% | estimated ± 0.5 pp, medium confidence |
| RealWorldQA | 86.5% | estimated ± 0.8 pp, medium confidence |
| Video-MME (with subtitle) | 87.9% | estimated ± 0.9 pp, medium confidence |
| OmniDocBench 1.5 | 91.7% | estimated ± 5.9 pp, low confidence |