Code Migration - CLI
Code · 2026-06-16
The CLI split of Code Migration. This child benchmark separates a source-reported subtask or subtrack from the parent aggregate so scores at different grains do not share one benchmark_slug.
Top models (higher is better)
| Model | Score |
|---|---|
| Claude Fable 5 | 60.1 |
| Claude Opus 5 | 53.3 |
| GPT-5.6 Sol | 47.2 |
| Opus 4.8 | 40.1 |
| GPT-5.5 | 36.9 |
| Sonnet 5 | 36.3 |
| GPT-5.6 Luna | 36.1 |
| Opus 4.7 | 35.2 |
| Sonnet 4.6 | 30.3 |
| GLM-5.2 | 27.2 |
| GPT-5.6 Terra | 26.6 |
| Grok 4.5 | 25.5 |
| GPT-5.4 | 23.8 |
| Gemini 3.6 Flash | 19.3 |
| GLM-5.1 | 18.8 |
| Muse Spark 1.1 | 18.6 |
| Kimi K2.6 | 18.0 |
| Gemini 3.5 Flash | 17.2 |
| DeepSeek-V4-Pro | 15.5 |
| Qwen3.6 Plus (2026-04-02) | 14.4 |
| MiMo-V2.5-Pro | 14.0 |
| Kimi K2.7 Code | 13.4 |
| MiniMax M3 | 11.9 |
| Gemini 3.1 Pro Preview | 10.8 |
| MiniMax M2.7 | 10.2 |
| Haiku 4.5 | 10.1 |
| MiMo-V2.5 | 9.9 |
| Grok 4.3 | 9.1 |
| Gemini 3 Flash Preview | 8.5 |
| GPT-5.4 Mini | 8.2 |
| Inkling | 7.0 |
| Gemini 3.5 Flash-Lite | 5.9 |
| Gemini 3.1 Flash-Lite Preview | 5.7 |