Agents' Last Exam ALE-CLI Pass Rate
Agents · 2026-06-03
Perfect-score pass rate on the 105-task ALE-CLI Linux-only subset of Agents' Last Exam.
Top models (higher is better)
| Model | Score |
|---|---|
| GPT-5.6 Luna | 28.6 |
| GPT-5.6 Sol | 28.6 |
| Kimi K3 | 27.6 |
| GPT-5.5 | 27.1 |
| GPT-5.6 Terra | 26.2 |
| Opus 4.8 | 25.7 |
| Claude Fable 5 | 23.8 |
| GLM-5.2 | 23.8 |
| GPT-5.4 | 22.5 |
| Opus 4.7 | 21.9 |
| Composer 2.5 | 19.0 |
| Doubao Seed 2.1 Pro | 19.0 |
| Gemini 3.1 Pro Preview | 17.1 |
| DeepSeek-V4-Pro | 14.1 |
| Sonnet 4.6 | 13.3 |
| Qwen3.7-Max | 13.3 |
| GLM-5.1 | 12.9 |
| Qwen3.6 Plus (2026-04-02) | 10.5 |
| MiMo-V2.5 | 10.0 |
| Kimi K2.6 | 8.1 |
| Grok 4.3 | 7.6 |
| MiniMax M2.7 | 5.7 |