Neura Intelligence Index
American Invitational Mathematics Examination 2025 — competition-level math problems.
| # | Model | Score | Provider | Source |
|---|---|---|---|---|
| 1 | 100 | verified | ||
| 2 | X Grok-4 Heavy🇺🇸 | 100 | xAI | verified |
| 3 | 100 | OpenAI | verified | |
| 4 | 100 | OpenAI | verified | |
| 5 | M Kimi K2-Thinking-0905🇨🇳 | 100 | Moonshot AI | verified |
| 6 | 99.8 | Anthropic | verified | |
| 7 | 99.7 | verified | ||
| 8 | 99.6 | OpenAI | verified | |
| 9 | M LongCat-Flash-Thinking-2601🇨🇳 | 99.6 | Meituan | verified |
| 10 | 99.2 | NVIDIA | verified | |
| 11 | 98.7 | OpenAI | verified | |
| 12 | 98.4 | OpenAI | verified | |
| 13 | B Seed 2.0 Pro🇨🇳 | 98.3 | ByteDance | verified |
| 14 | S Step-3.5-Flash🇨🇳 | 97.3 | StepFun | verified |
| 15 | 97 | Microsoft | verified | |
| 16 | S Sarvam-105B🇮🇳 | 96.7 | Sarvam AI | verified |
| 17 | S Sarvam-30B🇮🇳 | 96.7 | Sarvam AI | verified |
| 18 | 96.7 | OpenAI | verified | |
| 19 | M Kimi K2.5🇨🇳 | 96.1 | Moonshot AI | verified |
| 20 | 96 | DeepSeek | verified | |
| 21 | Z GLM-4.7🇨🇳 | 95.7 | Zhipu AI | verified |
| 22 | 94.6 | OpenAI | verified | |
| 23 | 94.6 | OpenAI | verified | |
| 24 | X MiMo-V2-Flash🇨🇳 | 94.1 | Xiaomi | verified |
| 25 | 94 | OpenAI | verified | |
| 26 | 94 | OpenAI | verified | |
| 27 | 94 | OpenAI | verified | |
| 28 | Z GLM-4.6🇨🇳 | 93.9 | Zhipu AI | verified |
| 29 | X Grok-3🇺🇸 | 93.3 | xAI | verified |
| 30 | 93.1 | DeepSeek | verified | |
| 31 | 93.1 | DeepSeek | verified | |
| 32 | B Seed 2.0 Lite🇨🇳 | 93 | ByteDance | verified |
| 33 | L K-EXAONE-236B-A23B🇰🇷 | 92.8 | LG AI Research | verified |
| 34 | 92.7 | OpenAI | verified | |
| 35 | 92.5 | OpenAI | verified | |
| 36 | 92.3 | Amazon | verified | |
| 37 | Q Qwen3-235B-A22B-Thinking-2507🇨🇳 | 92.3 | Alibaba Cloud / Qwen Team | verified |
| 38 | 92.1 | Amazon | verified | |
| 39 | X Grok 4 Fast🇺🇸 | 92 | xAI | verified |
| 40 | X Grok-4🇺🇸 | 91.7 | xAI | verified |
| 41 | Z GLM-4.7-Flash🇨🇳 | 91.6 | Zhipu AI | verified |
| 42 | I Mercury 2🇺🇸 | 91.1 | Inception | verified |
| 43 | 91.1 | OpenAI | verified | |
| 44 | 91 | Amazon | verified | |
| 45 | X Grok-3 Mini🇺🇸 | 90.8 | xAI | verified |
| 46 | M LongCat-Flash-Thinking🇨🇳 | 90.6 | Meituan | verified |
| 47 | 90.2 | NVIDIA | verified | |
| 48 | 90 | Cohere | verified | |
| 49 | Q Qwen3 VL 235B A22B Thinking🇨🇳 | 89.7 | Alibaba Cloud / Qwen Team | verified |
| 50 | 89.3 | DeepSeek | verified | |
| 51 | 88.9 | OpenAI | verified | |
| 52 | 88 | verified | ||
| 53 | Q Qwen3-Next-80B-A3B-Thinking🇨🇳 | 87.8 | Alibaba Cloud / Qwen Team | verified |
| 54 | S Step3-VL-10B🇨🇳 | 87.7 | StepFun | verified |
| 55 | 87.5 | DeepSeek | verified | |
| 56 | 87 | Anthropic | verified | |
| 57 | 87 | Baidu | verified | |
| 58 | 86.4 | OpenAI | verified | |
| 59 | 86.3 | Mistral AI | verified | |
| 60 | 85.2 | OpenAI | verified | |
| 61 | 85 | Mistral AI | verified | |
| 62 | 83.8 | Mistral AI | verified | |
| 63 | Q Qwen3 VL 32B Thinking🇨🇳 | 83.7 | Alibaba Cloud / Qwen Team | verified |
| 64 | Q Qwen3 VL 30B A3B Thinking🇨🇳 | 83.1 | Alibaba Cloud / Qwen Team | verified |
| 65 | 83 | verified | ||
| 66 | Q Qwen3 Max🇨🇳 | 81.6 | Alibaba Cloud / Qwen Team | verified |
| 67 | Q Qwen3 235B A22B🇨🇳 | 81.5 | Alibaba Cloud / Qwen Team | verified |
| 68 | 81.2 | OpenAI | verified | |
| 69 | 81 | MiniMax | verified | |
| 70 | 80.7 | Anthropic | verified | |
| 71 | Q Qwen3 VL 8B Thinking🇨🇳 | 80.3 | Alibaba Cloud / Qwen Team | verified |
| 72 | 78.7 | Mistral AI | verified | |
| 73 | O MiniCPM-SALA🇨🇳 | 78.3 | OpenBMB | verified |
| 74 | 78 | MiniMax | verified | |
| 75 | 78 | Anthropic | verified | |
| 76 | 78 | Microsoft | verified | |
| 77 | 76.9 | MiniMax | verified | |
| 78 | 75.5 | Anthropic | verified | |
| 79 | Q Qwen3 VL 235B A22B Instruct🇨🇳 | 74.7 | Alibaba Cloud / Qwen Team | verified |
| 80 | 74.6 | MiniMax | verified | |
| 81 | Q Qwen3 VL 4B Thinking🇨🇳 | 74.5 | Alibaba Cloud / Qwen Team | verified |
| 82 | Q Qwen3 32B🇨🇳 | 72.9 | Alibaba Cloud / Qwen Team | verified |
| 83 | 72.5 | NVIDIA | verified | |
| 84 | 72.1 | NVIDIA | verified | |
| 85 | 72.1 | Mistral AI | verified | |
| 86 | 72 | verified | ||
| 87 | Q Qwen3 30B A3B🇨🇳 | 70.9 | Alibaba Cloud / Qwen Team | verified |
| 88 | 70.5 | Anthropic | verified | |
| 89 | Q Qwen3-235B-A22B-Instruct-2507🇨🇳 | 70.3 | Alibaba Cloud / Qwen Team | verified |
| 90 | Q Qwen3-Next-80B-A3B-Instruct🇨🇳 | 69.5 | Alibaba Cloud / Qwen Team | verified |
| 91 | Q Qwen3 VL 30B A3B Instruct🇨🇳 | 69.3 | Alibaba Cloud / Qwen Team | verified |
| 92 | Q Qwen3 VL 32B Instruct🇨🇳 | 66.2 | Alibaba Cloud / Qwen Team | verified |
| 93 | 64.9 | Mistral AI | verified | |
| 94 | M LongCat-Flash-Lite🇨🇳 | 63.2 | Meituan | verified |
| 95 | 62.9 | Microsoft | verified | |
| 96 | 62.8 | Mistral AI | verified | |
| 97 | M LongCat-Flash-Chat🇨🇳 | 61.3 | Meituan | verified |
| 98 | 58.4 | NVIDIA | verified | |
| 99 | 54.8 | Anthropic | verified | |
| 100 | 49.8 | verified | ||
| 101 | 49.8 | DeepSeek | verified | |
| 102 | M Kimi K2 Instruct🇨🇳 | 49.5 | Moonshot AI | verified |
| 103 | M Kimi K2-Instruct-0905🇨🇳 | 49.5 | Moonshot AI | verified |
| 104 | 47.1 | NVIDIA | verified | |
| 105 | Q Qwen3 VL 4B Instruct🇨🇳 | 46.6 | Alibaba Cloud / Qwen Team | verified |
| 106 | 46.4 | OpenAI | verified | |
| 107 | Q Qwen3 VL 8B Instruct🇨🇳 | 45.9 | Alibaba Cloud / Qwen Team | verified |
| 108 | 42.1 | OpenAI | verified | |
| 109 | 40.2 | OpenAI | verified | |
| 110 | 23.3 | verified | ||
| 111 | 11.6 | verified | ||
| 112 | 11.6 | verified | ||
| 113 | 6.7 | verified | ||
| 114 | 6.7 | verified |