Neura Intelligence Index
Back to Leaderboard
TerminalBench
Terminal/CLI task completion — executing shell commands to achieve goals.
coding0-100 percent
Model Rankings (25)
| # | Model | Score | Provider | Source |
|---|---|---|---|---|
| 1 | 50 | Anthropic | verified | |
| 2 | 47.9 | MiniMax | verified | |
| 3 | M Kimi K2-Thinking-0905🇨🇳 | 47.1 | Moonshot AI | verified |
| 4 | 46.3 | MiniMax | verified | |
| 5 | 43.3 | Anthropic | verified | |
| 6 |  Nova 2 Pro🇺🇸 | 41.3 | Amazon | verified |
| 7 | 41 | Anthropic | verified | |
| 8 | Z GLM-4.6🇨🇳 | 40.5 | Zhipu AI | verified |
| 9 | M LongCat-Flash-Chat🇨🇳 | 39.5 | Meituan | verified |
| 10 | 39.2 | Anthropic | verified | |
| 11 | 37.7 | DeepSeek | verified | |
| 12 | Z GLM-4.5🇨🇳 | 37.5 | Zhipu AI | verified |
| 13 | 35.5 | Anthropic | verified | |
| 14 | 35.2 | Anthropic | verified | |
| 15 | M LongCat-Flash-Lite🇨🇳 | 33.8 | Meituan | verified |
| 16 | Z GLM-4.7🇨🇳 | 33.3 | Zhipu AI | verified |
| 17 | Nova 2 Lite🇺🇸 | 32.5 | Amazon | verified |
| 18 | 31.3 | DeepSeek | verified | |
| 19 | X MiMo-V2-Flash🇨🇳 | 30.5 | Xiaomi | verified |
| 20 | Z GLM-4.5-Air🇨🇳 | 30 | Zhipu AI | verified |
| 21 | M Kimi K2 Instruct🇨🇳 | 30 | Moonshot AI | verified |
| 22 | 25.8 | NVIDIA | verified | |
| 23 | M Kimi K2-Instruct-0905🇨🇳 | 25 | Moonshot AI | verified |
| 24 | 8.5 | NVIDIA | verified | |
| 25 | 5.7 | DeepSeek | verified |