Agents' Last Exam ALE-CLI Score
Agents · 2026-06-03
Average partial-credit rubric score on the 105-task ALE-CLI Linux-only subset of Agents' Last Exam.
Top models (higher is better)
| Model | Score |
|---|---|
| GPT-5.6 Sol | 53.6 |
| Kimi K3 | 52.8 |
| GPT-5.6 Luna | 50.7 |
| GPT-5.6 Terra | 49.8 |
| GPT-5.5 | 49.2 |
| Claude Fable 5 | 48.5 |
| Opus 4.7 | 44.7 |
| Opus 4.8 | 44.6 |
| GLM-5.2 | 43.4 |
| GPT-5.4 | 41.8 |
| Doubao Seed 2.1 Pro | 41.4 |
| Composer 2.5 | 40.8 |
| Gemini 3.1 Pro Preview | 36.2 |
| Qwen3.7-Max | 34.0 |
| Sonnet 4.6 | 32.0 |
| DeepSeek-V4-Pro | 30.8 |
| GLM-5.1 | 30.8 |
| Qwen3.6 Plus (2026-04-02) | 28.6 |
| MiMo-V2.5 | 26.5 |
| Grok 4.3 | 24.3 |
| Kimi K2.6 | 21.2 |
| MiniMax M2.7 | 14.6 |