BLXBench Debugging category
Code · 2026-04-25
Debugging category component of the BLXBench public leaderboard. The source describes this category as bug fixes, edge conditions, and minimal patch accuracy; scores are category score percentages.
Top models (higher is better)
| Model | Score |
|---|---|
| Qwen3.7-Max | 80.2 |
| Kimi K2.6 | 79.9 |
| GPT-5.3-Codex | 79.6 |
| GPT-5.5 | 77.0 |
| MiMo-V2.5 | 76.5 |
| DeepSeek-V4-Pro | 76.2 |
| GLM-5.1 | 75.8 |
| Grok Build 0.1 | 74.7 |
| CoBuddy | 73.9 |
| MiMo-V2.5-Pro | 73.8 |
| North Mini Code | 72.9 |
| DeepSeek-V4-Flash | 71.9 |
| Grok 4.3 | 71.2 |
| Nemotron 3 Super 120B A12B | 71.2 |
| Qwen3.7-Plus | 71.0 |
| MiniMax M3 | 69.9 |
| GLM-5.2 | 69.7 |
| Mistral Medium 3.5 | 69.6 |
| Qwen3.6-Flash-2026-04-16 | 69.0 |
| Gemini 3.1 Flash-Lite | 68.1 |
| Mistral Small 4 | 67.9 |
| Nemotron 3 Nano 30B A3B | 67.9 |
| Granite 4.1 8B | 67.2 |
| Step 3.7 Flash | 62.4 |
| Nemotron 3 Nano Omni 30B A3B | 61.9 |
| Claude Fable 5 | 58.2 |
| Kimi K2.7 Code | 53.1 |
| Ring 2.6 1T | 50.8 |
| MiniMax M2.7 | 45.5 |
| Opus 4.7 | 45.1 |
| Opus 4.8 | 41.3 |
| Gemini 3.5 Flash | 25.8 |