模型排行
| 排名 | 模型 | 机构 | 分数 | 置信区间 | 开放状态 | 证据 |
|---|---|---|---|---|---|---|
| 1 |  ornith-ai/Ornith-1.5-397B Resolved % ↑ | Ornith | 86 | Mixed / see source | 未验证 来源 | |
| 2 | mindlab-research/Macaron-V1-Venti Resolved % ↑ | Mind Lab | 85.60 | Mixed / see source | 未验证 来源 | |
| 3 | mindlab-research/Macaron-V1-Coding-Venti Resolved % ↑ | Mind Lab | 85.60 | Mixed / see source | 未验证 来源 | |
| 4 | ornith-ai/Ornith-1.0-397B Resolved % ↑ | Ornith | 82.40 | Mixed / see source | 未验证 来源 | |
| 5 | deepseek-ai/DeepSeek-V4-Pro Resolved % ↑ | DeepSeek | 80.60 | Mixed / see source | 未验证 来源 | |
| 6 | MiniMaxAI/MiniMax-M3 Resolved % ↑ | MiniMax | 80.50 | Mixed / see source | 未验证 来源 | |
| 7 | moonshotai/Kimi-K2.6 Resolved % ↑ | Moonshot AI | 80.20 | Mixed / see source | 未验证 来源 | |
| 8 | thinkingmachines/Inkling-Small Resolved % ↑ | Thinking Machines Lab | 80.20 | Mixed / see source | 未验证 来源 | |
| 9 | deepseek-ai/DeepSeek-V4-Flash Resolved % ↑ | DeepSeek | 79 | Mixed / see source | 未验证 来源 | |
| 10 | ornith-ai/Ornith-1.5-35B-A3B Resolved % ↑ | Ornith | 79 | Mixed / see source | 未验证 来源 | |
| 11 | XiaomiMiMo/MiMo-V2.5-Pro Resolved % ↑ | Xiaomi MiMo | 78.90 | Mixed / see source | 未验证 来源 | |
| 12 | dots-studio/dots3-note-prev Resolved % ↑ | dots studio | 78.40 | Mixed / see source | 未验证 来源 | |
| 13 | tencent/Hy3 Resolved % ↑ | Tencent | 78 | Mixed / see source | 未验证 来源 | |
| 14 | zai-org/GLM-5 Resolved % ↑ | Z.ai | 77.80 | Mixed / see source | 未验证 来源 | |
| 15 | mistralai/Mistral-Medium-3.5-128B Resolved % ↑ | Mistral AI_ | 77.60 | Mixed / see source | 未验证 来源 | |
| 16 | thinkingmachines/Inkling Resolved % ↑ | Thinking Machines Lab | 77.60 | Mixed / see source | 未验证 来源 | |
| 17 | Qwen/Qwen3.6-27B Resolved % ↑ | Qwen | 77.20 | Mixed / see source | 未验证 来源 | |
| 18 | Qwen/Qwen3.5-397B-A17B Resolved % ↑ | Qwen | 76.40 | Mixed / see source | 未验证 来源 | |
| 19 | meta-models/Muse-Glimmer-30B Resolved % ↑ | Meta Inc. | 76 | Mixed / see source | 未验证 来源 | |
| 20 | MiniMaxAI/MiniMax-M2.5 Resolved % ↑ | MiniMax | 75.80 | Mixed / see source | 未验证 来源 | |
| 21 | ornith-ai/Ornith-1.0-35B Resolved % ↑ | Ornith | 75.60 | Mixed / see source | 未验证 来源 | |
| 22 | mindlab-research/Macaron-V1-Tall Resolved % ↑ | Mind Lab | 75.40 | Mixed / see source | 未验证 来源 | |
| 23 | Multilingual-Multimodal-NLP/IndustrialCoder Resolved % ↑ | Multilingual-Multimodal-NLP | 74.80 | Mixed / see source | 未验证 来源 | |
| 24 | poolside/Laguna-M.1 Resolved % ↑ | Poolside | 74.60 | Mixed / see source | 未验证 来源 | |
| 25 | stepfun-ai/Step-3.5-Flash Resolved % ↑ | StepFun | 74.40 | Mixed / see source | 未验证 来源 | |
| 26 | tencent/Hy3-preview Resolved % ↑ | Tencent | 74.40 | Mixed / see source | 未验证 来源 | |
| 27 | MiniMaxAI/MiniMax-M2.1 Resolved % ↑ | MiniMax | 74 | Mixed / see source | 未验证 来源 | |
| 28 | inclusionAI/Ring-2.6-1T Resolved % ↑ | inclusionAI | 74 | Mixed / see source | 未验证 来源 | |
| 29 | zai-org/GLM-4.7 Resolved % ↑ | Z.ai | 73.80 | Mixed / see source | 未验证 来源 | |
| 30 | Qwen/Qwen3.6-35B-A3B Resolved % ↑ | Qwen | 73.40 | Mixed / see source | 未验证 来源 | |
| 31 | zai-org/GLM-5 Resolved % ↑ | Z.ai | 72.80 | Mixed / see source | 未验证 来源 | |
| 32 | Qwen/Qwen3.5-27B Resolved % ↑ | Qwen | 72.40 | Mixed / see source | 未验证 来源 | |
| 33 | inclusionAI/Ling-2.6-1T Resolved % ↑ | inclusionAI | 72.20 | Mixed / see source | 未验证 来源 | |
| 34 | Qwen/Qwen3.5-122B-A10B Resolved % ↑ | Qwen | 72 | Mixed / see source | 未验证 来源 | |
| 35 |  nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16 Resolved % ↑ | NVIDIA | 71.90 | Mixed / see source | 未验证 来源 | |
| 36 | moonshotai/Kimi-K2-Thinking Resolved % ↑ | Moonshot AI | 71.30 | Mixed / see source | 未验证 来源 | |
| 37 | poolside/Laguna-XS-2.1 Resolved % ↑ | Poolside | 70.90 | Mixed / see source | 未验证 来源 | |
| 38 | moonshotai/Kimi-K2.5 Resolved % ↑ | Moonshot AI | 70.80 | Mixed / see source | 未验证 来源 | |
| 39 | Qwen/Qwen3-Coder-Next Resolved % ↑ | Qwen | 70.60 | Mixed / see source | 未验证 来源 | |
| 40 | ornith-ai/Ornith-1.5-9B Resolved % ↑ | Ornith | 70.60 | Mixed / see source | 未验证 来源 | |
| 41 | upstage/Solar-Open2-250B Resolved % ↑ | upstage | 70.40 | Mixed / see source | 未验证 来源 | |
| 42 | poolside/Laguna-XS.2 Resolved % ↑ | Poolside | 69.90 | Mixed / see source | 未验证 来源 | |
| 43 | nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4 Resolved % ↑ | NVIDIA | 69.70 | Mixed / see source | 未验证 来源 | |
| 44 | MiniMaxAI/MiniMax-M2 Resolved % ↑ | MiniMax | 69.40 | Mixed / see source | 未验证 来源 | |
| 45 | ornith-ai/Ornith-1.0-9B Resolved % ↑ | Ornith | 69.40 | Mixed / see source | 未验证 来源 | |
| 46 | MuVeraAI/Laguna-XS.2 Resolved % ↑ | MuVeraAI Corporation | 68.20 | Mixed / see source | 未验证 来源 | |
| 47 | CohereLabs/North-Mini-Code-1.0 Resolved % ↑ | Cohere Labs | 67.60 | Mixed / see source | 未验证 来源 | |
| 48 | internlm/Intern-S2-Preview Resolved % ↑ | Intern Large Models | 64 | Mixed / see source | 未验证 来源 | |
| 49 | openpangu/openPangu-2.0-Flash Resolved % ↑ | openPangu | 63.10 | Mixed / see source | 未验证 来源 | |
| 50 |  openai/gpt-oss-120b Resolved % ↑ | OpenAI | 62.40 | Mixed / see source | 未验证 来源 |