Seed1.6-Thinking ranks first on MathCanvas-Bench.
ByteDance Seed Jun 2025 Closed weights Flagship
Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| MMLU-Pro | 86.6% | #21 of 212 | Official leaderboard | |
| WritingBench | 78.8 | #11 of 45 | Official leaderboard |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| AIME 2024 | 90% | #19 of 88 | Lab model card | |
| GPQA Diamond | 81.5% | #106 of 302 | Lab model card | |
| AIME 2025 | 86% | #69 of 143 | Lab model card | |
| Beyond AIME | 56.3% | #11 of 14 | Lab model card |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| LiveCodeBench | 64% | #19 of 54 | Lab model card | |
| AutoCodeBench | 44.7% | #17 of 34 | Official leaderboard |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| MathCanvas-Bench | 44.1% | #1 of 16 | Official leaderboard | |
| MMMU | 74.8% | #22 of 136 | Official leaderboard | |
| InfographicVQA | 88% | #6 of 26 | Lab model card | |
| Math-VR | 58.4% | #6 of 26 | Official leaderboard | |
| ChartQA | 89.1% | #10 of 37 | Lab model card | |
| RealWorldQA | 77.8% | #20 of 38 | Lab model card | |
| MATH-Vision (MathVision) | 70.3% | #32 of 51 | Lab model card | |
| MMMU-Pro | 66.4% | #57 of 90 | Lab model card | |
| VisFactor | 21.5% | #17 of 26 | Official leaderboard |
The Self-Reported Index fits one score to the 11 capability results ByteDance Seed reported for this model, on the scale of the Epoch Capabilities Index. Labs choose which benchmarks to show and how to run them, so read it as the lab's claim, not an independent measurement. The 90% interval runs from 136.4 to 140.6. How we compute it.
Where the benchmark's own leaderboard, or Epoch AI's run, also has this model. Gaps mostly come from different harnesses, effort levels and task subsets. Every pair.
The benchmarks the lab chose to headline, with the scores it gave.