MLX Benchmark V2 - Debugging: leaderboard
MLX Benchmark V2 debugging slice for identifying and fixing Apple MLX code issues.
Metric: Accuracy (%). Source: huggingface.co. Saturation forecast: Not forecast. 21 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Sonnet 4.6 | 95.24 | #90 |
| 2 | Gemini 3 Flash (Preview) | 80.95 | #82 |
| 3 | Grok 4.1 Fast | 71.43 | #217 |
| 4 | Gemma 4 26B A4B (IT) | 66.67 | #297 |
| 5 | GPT-5.4 Nano | 57.14 | #328 |
| 6 | Gemini 2.5 Flash Lite (Preview 09-2025) | 52.38 | #318 |
| 7 | DeepSeek V4 Pro | 33.33 | #102 |
| 8 | Qwen 3.6 35B A3B | 33.33 | #226 |
| 9 | Nemotron 3 Ultra | 19.05 | #235 |
| 10 | GPT-5 Nano | 14.29 | #425 |
| 11 | DeepSeek V4 Flash | 14.29 | #137 |
| 12 | Kimi K2.5 | 0 | #144 |
| 13 | Kimi K2.6 | 0 | #104 |
| 14 | GLM-5.1 | 0 | #97 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=mlx-benchmark-v2-debugging · How It Works · Data refreshed daily, snapshot 2026-10-20.