| 1 | MAI-Thinking-1 | Microsoft | Proprietary | 85% | |
| 2 | Qwen3.8 Max | Alibaba | Open weights | 82.8% | |
| 3 | Inkling-Small | Thinking Machines Lab | Open weights | 82.2% | |
| 4 | Nemotron 3 Ultra | NVIDIA · Open weight | 81.7% | | |
| 5 | Qwen3.8-Omni-Flash | Alibaba · Closed | 81.5% | | |
| 6 | Grok 4.3 | xAI · Closed | 81.3% | | |
| 7 | Qwen3.8-Flash-Next | Alibaba · Open weight | 81.3% | | |
| 8 | dots3-note Preview | Dots Studio · Open weight | 80.4% | | |
| 9 | Solar Open 2 | Upstage · Open weight | 80% | | |
| 10 | Inkling | Thinking Machines Lab · Open weight | 79.8% | | |
| 11 | Qwen3.8-27B | Alibaba · Open weight | 79.5% | | |
| 12 | Granite 4.2 8B | IBM · Open weight | 79.3% | | |
| 13 | Qwen3.7 Max | Alibaba · Closed | 79.1% | | |
| 14 | Qwen3.7 Plus | Alibaba · Closed | 79.1% | | |
| 15 | Granite 4.2 30B | IBM · Open weight | 77.2% | | |
| 16 | Muse Glimmer 30B | Meta · Open weight | 77% | | |
| 17 | Mercury 2.5 | Inception · Closed | 77% | | |
| 18 | Gemini 3.5 Flash | Google · Closed | 76.3% | | |
| 19 | A.X K2 | SK Telecom · Open weight | 75.9% | | |
| 20 | Qwen3.6 Plus | Alibaba · Closed | 75.8% | | |
| 21 | Ling 3.0 Flash | InclusionAI · Open weight | 74.5% | | |
| 22 | Granite 4.2 3B | IBM · Open weight | 74.3% | | |
| 23 | Nemotron 3 Nano Omni 30B A3B | NVIDIA · Open weight | 74.2% | | |
| 24 | Ternary Bonsai 2 27B | Prism ML · Open weight | 74% | | |
| 25 | Ling 3.0 Flash FP8 | InclusionAI · Open weight | 73.4% | | |
| 26 | Nemotron 3.5 Lightning 30B A3B NVFP4 | NVIDIA · Open weight | 72.9% | | |
| 27 | K-EXAONE 2.0 | LG AI Research · Open weight | 72.6% | | |
| 28 | Mercury 2 | Inception · Closed | 71.3% | | |
| 29 | Agents-A1-4B | InternScience · Open weight | 69.1% | | |
| 30 | MiniCPM5-2B | OpenBMB · Open weight | 66.3% | | |
| 31 | Hy3 Preview | Tencent · Open weight | 63.1% | | |
| 32 | LFM2.5-2.6B | LiquidAI · Open weight | 59.2% | | |
| 33 | Claude Opus 4.5 | Anthropic · Closed | 58% | | |
| 34 | Ling 2.6 Flash | InclusionAI · Open weight | 57% | | |
| 35 | LFM2.5-8B-A1B | LiquidAI · Open weight | 56.5% | | |
| 36 | Solar Pro 3 | Upstage · Closed | 55.8% | | |
| 37 | ZAYA1-8B | Zyphra · Open weight | 52.6% | | |
| 38 | MiniCPM5-1B | OpenBMB · Open weight | 46.7% | | |
| 39 | LFM2.5-230M | LiquidAI · Open weight | 38.4% | | |
| 40 | Kanana-2 1.3B Instruct | Kakao · Open weight | 34.7% | | |
| 41 | Kanana-2 3B Instruct | Kakao · Open weight | 33.3% | | |
| 42 | LFM2.5-VL-3B | LiquidAI · Open weight | 25.8% | | |
| 43 | LLaDA2.2-mini | InclusionAI · Open weight | 24.9% | | |