seed-oss-36b-instruct.
ByteDance Seed Aug 2025 Open weights Mid-size
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 | |
|---|---|---|---|---|
| RULER | 94.6% | #10 of 38 | Model card | |
| MMLU | 87.4% | #36 of 127 | Model card | |
| MMLU-Pro | 82.7% | #49 of 151 | Model card | |
| IFEval | 85.8% | #69 of 126 | Model card | |
| MMMLU | 78.4% | #44 of 62 | Model card | |
| SuperGPQA | 55.7% | #21 of 26 | Model card | |
| SimpleQA | 9.70% | #50 of 59 | Model card | |
| Humanity's Last Exam | 10.1% | #135 of 147 | Model card |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| AIME 2024 | 91.7% | #11 of 87 | Model card | |
| GPQA Diamond | 71.5% | #106 of 219 | Epoch AI | |
| OTIS Mock AIME 2024-2025 | 67.5% | #105 of 202 | Epoch AI | |
| AIME 2025 | 84.7% | #75 of 143 | Model card | |
| Beyond AIME | 65% | #8 of 14 | Model card | |
| KOR-Bench | 70.6% | #3 of 4 | Model card | |
| ARC-AGI-2 | 1.45% | #22 of 22 | Model card |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| LiveCodeBench v6 | 67.4% | #55 of 76 | Model card | |
| SWE-bench Verified | 56% | #116 of 150 | Model card | |
| Multi-SWE-bench | 17% | #10 of 10 | Model card |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| tau-bench | 58.2% | #14 of 23 | Model card |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| AIR-Bench 2024 | 75.6% | #2 of 2 | Model card |
The benchmarks the lab chose to headline, with the scores it gave.