Neura Intelligence Index
Back to Leaderboard
Humanity's Last Exam
Expert-level questions across diverse domains, designed to be the hardest public benchmark.
reasoning0-100 percent
Model Rankings (90)
| # | Model | Score | Provider | Source |
|---|---|---|---|---|
| 1 | 64.7 | Anthropic | verified | |
| 2 | 64.5 | Anthropic | verified | |
| 3 | M Muse Spark 1.1🇺🇸 | 62.1 | Meta | verified |
| 4 | M Muse Spark🇺🇸 | 58.4 | Meta | verified |
| 5 | 57.9 | Anthropic | verified | |
| 6 | 57.4 | Anthropic | verified | |
| 7 | 57.2 | OpenAI | verified | |
| 8 | M Kimi K3🇨🇳 | 56 | Moonshot AI | verified |
| 9 | B Seed 2.1 Pro🇨🇳 | 55.7 | ByteDance | verified |
| 10 | 54.7 | Anthropic | verified | |
| 11 | Z GLM-5.2🇨🇳 | 54.7 | Zhipu AI | verified |
| 12 | B Seed 2.1 Turbo🇨🇳 | 54.6 | ByteDance | verified |
| 13 | 53.1 | Anthropic | verified | |
| 14 | Z GLM-5.1🇨🇳 | 52.3 | Zhipu AI | verified |
| 15 | 52.2 | OpenAI | verified | |
| 16 | 51.4 | verified | ||
| 17 | M Kimi K2-Thinking-0905🇨🇳 | 51 | Moonshot AI | verified |
| 18 | X Grok-4 Heavy🇺🇸 | 50.7 | xAI | verified |
| 19 | M Kimi K2.5🇨🇳 | 50.2 | Moonshot AI | verified |
| 20 | 49 | Anthropic | verified | |
| 21 | Q Qwen3.5-27B🇨🇳 | 48.5 | Alibaba Cloud / Qwen Team | verified |
| 22 | 48.2 | DeepSeek | verified | |
| 23 | Q Qwen3.5-122B-A10B🇨🇳 | 47.5 | Alibaba Cloud / Qwen Team | verified |
| 24 | Q Qwen3.5-35B-A3B🇨🇳 | 47.4 | Alibaba Cloud / Qwen Team | verified |
| 25 | 45.8 | verified | ||
| 26 | 45.1 | DeepSeek | verified | |
| 27 | 43.5 | verified | ||
| 28 | Z GLM-4.7🇨🇳 | 42.8 | Zhipu AI | verified |
| 29 | Q Qwen3.7 Max🇨🇳 | 41.4 | Alibaba Cloud / Qwen Team | verified |
| 30 | 40.8 | DeepSeek | verified | |
| 31 | 40.2 | verified | ||
| 32 | X Grok-4🇺🇸 | 40 | xAI | verified |
| 33 | 39.8 | OpenAI | verified | |
| 34 | 39 | Baidu | verified | |
| 35 | 37.4 | NVIDIA | verified | |
| 36 | 36.6 | OpenAI | verified | |
| 37 | M Kimi K2.6🇨🇳 | 36.4 | Moonshot AI | verified |
| 38 | Q Qwen3.7-Plus🇨🇳 | 34.7 | Alibaba Cloud / Qwen Team | verified |
| 39 | 34.5 | OpenAI | verified | |
| 40 | X MiMo-V2.5-Pro🇨🇳 | 34 | Xiaomi | verified |
| 41 | 30.6 | DeepSeek | verified | |
| 42 | Q Qwen3.6 Plus🇨🇳 | 28.8 | Alibaba Cloud / Qwen Team | verified |
| 43 | Q Qwen3.5-397B-A17B🇨🇳 | 28.7 | Alibaba Cloud / Qwen Team | verified |
| 44 | 28.2 | OpenAI | verified | |
| 45 | 26.5 | verified | ||
| 46 | M LongCat-Flash-Thinking-2601🇨🇳 | 25.2 | Meituan | verified |
| 47 | 25.1 | DeepSeek | verified | |
| 48 | 24.8 | OpenAI | verified | |
| 49 | 24.3 | OpenAI | verified | |
| 50 | Q Qwen3.6-27B🇨🇳 | 24 | Alibaba Cloud / Qwen Team | verified |
| 51 | 22.8 | NVIDIA | verified | |
| 52 | X MiMo-V2-Flash🇨🇳 | 22.1 | Xiaomi | verified |
| 53 | 22 | MiniMax | verified | |
| 54 | 21.6 | verified | ||
| 55 | Q Qwen3.6-35B-A3B🇨🇳 | 21.4 | Alibaba Cloud / Qwen Team | verified |
| 56 | X Grok 4 Fast🇺🇸 | 20 | xAI | verified |
| 57 | 19.8 | DeepSeek | verified | |
| 58 | Q Qwen3-235B-A22B-Thinking-2507🇨🇳 | 18.2 | Alibaba Cloud / Qwen Team | verified |
| 59 | 18 | Microsoft | verified | |
| 60 | 17.8 | verified | ||
| 61 | 17.7 | DeepSeek | verified | |
| 62 | 17.2 | verified | ||
| 63 | Z GLM-4.6🇨🇳 | 17.2 | Zhipu AI | verified |
| 64 | 16.7 | OpenAI | verified | |
| 65 | 16 | verified | ||
| 66 | 15.9 | DeepSeek | verified | |
| 67 | 15.5 | NVIDIA | verified | |
| 68 | 14.9 | OpenAI | verified | |
| 69 | 14.7 | OpenAI | verified | |
| 70 | 14.7 | OpenAI | verified | |
| 71 | Z GLM-4.5🇨🇳 | 14.4 | Zhipu AI | verified |
| 72 | Z GLM-4.7-Flash🇨🇳 | 14.4 | Zhipu AI | verified |
| 73 | Q Qwen3 VL 235B A22B Thinking🇨🇳 | 13.6 | Alibaba Cloud / Qwen Team | verified |
| 74 | 12.5 | MiniMax | verified | |
| 75 | 11.9 | verified | ||
| 76 | S Sarvam-105B🇮🇳 | 11.2 | Sarvam AI | verified |
| 77 | 11 | verified | ||
| 78 | 10.9 | OpenAI | verified | |
| 79 | Z GLM-4.5-Air🇨🇳 | 10.6 | Zhipu AI | verified |
| 80 | 9 | Mistral AI | verified | |
| 81 | 8.7 | OpenAI | verified | |
| 82 | 8.4 | MiniMax | verified | |
| 83 | 7.2 | MiniMax | verified | |
| 84 | 5.4 | OpenAI | verified | |
| 85 | 5.3 | OpenAI | verified | |
| 86 | 5.2 | verified | ||
| 87 | 5.1 | verified | ||
| 88 | M Kimi K2 Instruct🇨🇳 | 4.7 | Moonshot AI | verified |
| 89 | M Kimi K2-Instruct-0905🇨🇳 | 4.7 | Moonshot AI | verified |
| 90 | 3.7 | OpenAI | verified |