Hunyuan-A13B-Instruct places in the top three on 2 benchmarks.
Tencent Jun 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 | |
|---|---|---|---|---|
| BIG-Bench Hard | 89.1% | #7 of 71 | Model card | |
| DROP | 91.1% | #7 of 40 | Model card | |
| MMLU-Pro | 67.3% | #94 of 212 | Official leaderboard | |
| IFEval | 84.7% | #78 of 126 | Model card | |
| FRAMES | 81.1% | #5 of 8 | Model card | |
| LongBench v2 | 55% | #17 of 27 | Model card | |
| SysBench | 76.1% | #4 of 6 | Model card | |
| RULER | 76.7% | #33 of 38 | Model card | |
| ComplexNLU (Tencent) | 61.2% | #1 of 1 | Model card | |
| InsCtrl (Tencent) | 71.9% | #1 of 1 | Model card | |
| LengthCtrl (Tencent) | 55.4% | #1 of 1 | Model card | |
| PenguinScrolls | 87.7% | #1 of 1 | Model card | |
| Word-Task (Tencent) | 62.9% | #1 of 1 | Model card |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| MATH | 94.3% | #5 of 93 | Model card | |
| AIME 2024 | 87.3% | #27 of 87 | Model card | |
| OlympiadBench | 82.7% | #1 of 3 | Model card | |
| GPQA Diamond | 71.2% | #163 of 284 | Model card | |
| ZebraLogic | 84.7% | #14 of 23 | Model card | |
| AIME 2025 | 76.8% | #100 of 143 | Model card |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| FullStackBench | 67.8% | #2 of 6 | Model card | |
| LiveCodeBench | 63.9% | #31 of 73 | Model card | |
| ArtifactsBench | 43% | #4 of 5 | Model card | |
| McEval | 59.6% | #1 of 1 | Model card |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Berkeley Function Calling Leaderboard (BFCL) | 78.3% | #3 of 47 | Model card | |
| ComplexFuncBench | 61.2% | #1 of 3 | Model card | |
| tau-bench | 54.7% | #20 of 23 | Model card | |
| SnakeBench | 15 | #164 of 183 | Official leaderboard | |
| C3-Bench | 63.5% | #1 of 1 | Model card |
The benchmarks the lab chose to headline, with the scores it gave.
Every result in the longer documents the lab published with it. 5 of them aren't in the launch post.