Neura Intelligence Index
Harder subset of SWE-Bench with more complex multi-file changes.
| # | Model | Score | Provider | Source |
|---|---|---|---|---|
| 1 | 80 | Anthropic | verified | |
| 2 | 77.8 | Anthropic | verified | |
| 3 | 69.2 | Anthropic | verified | |
| 4 | X Grok 4.5πΊπΈ | 64.7 | xAI | verified |
| 5 | 64.6 | OpenAI | verified | |
| 6 | 64.3 | Anthropic | verified | |
| 7 | 63.4 | OpenAI | verified | |
| 8 | 63.2 | Anthropic | verified | |
| 9 | 62.7 | OpenAI | verified | |
| 10 | Z GLM-5.2π¨π³ | 62.1 | Zhipu AI | verified |
| 11 | M Muse Spark 1.1πΊπΈ | 61.5 | Meta | verified |
| 12 | Q Qwen3.7 Maxπ¨π³ | 60.6 | Alibaba Cloud / Qwen Team | verified |
| 13 | 59 | MiniMax | verified | |
| 14 | 58.6 | OpenAI | verified | |
| 15 | M Kimi K2.6π¨π³ | 58.6 | Moonshot AI | verified |
| 16 | Z GLM-5.1π¨π³ | 58.4 | Zhipu AI | verified |
| 17 | T Hy3π¨π³ | 57.9 | Tencent | verified |
| 18 | 57.7 | OpenAI | verified | |
| 19 | Q Qwen3.7-Plusπ¨π³ | 57.6 | Alibaba Cloud / Qwen Team | verified |
| 20 | B Seed 2.1 Proπ¨π³ | 57.5 | ByteDance | verified |
| 21 | X MiMo-V2.5-Proπ¨π³ | 57.2 | Xiaomi | verified |
| 22 | B Seed 2.1 Turboπ¨π³ | 57 | ByteDance | verified |
| 23 | 56.8 | OpenAI | verified | |
| 24 | Q Qwen3.6 Plusπ¨π³ | 56.6 | Alibaba Cloud / Qwen Team | verified |
| 25 | 56.4 | OpenAI | verified | |
| 26 | 56.2 | MiniMax | verified | |
| 27 | X MiMo-V2.5π¨π³ | 56.1 | Xiaomi | verified |
| 28 | 55.4 | DeepSeek | verified | |
| 29 | 55.4 | MiniMax | verified | |
| 30 | 55.1 | verified | ||
| 31 | 54.4 | OpenAI | verified | |
| 32 | 54.2 | verified | ||
| 33 | Q Qwen3.6-27Bπ¨π³ | 53.5 | Alibaba Cloud / Qwen Team | verified |
| 34 | 52.8 | Microsoft | verified | |
| 35 | 52.6 | DeepSeek | verified | |
| 36 | M Muse SparkπΊπΈ | 52.4 | Meta | verified |
| 37 | 52.4 | OpenAI | verified | |
| 38 | 51.2 | Microsoft | verified | |
| 39 | M Kimi K2.5π¨π³ | 50.7 | Moonshot AI | verified |
| 40 | Q Qwen3.6-35B-A3Bπ¨π³ | 49.5 | Alibaba Cloud / Qwen Team | verified |
| 41 | 40.2 | Cohere | verified |