AI Coding Daily (OpenCode) - Python Code Quality: leaderboard
AI Coding Daily native Python Code Quality (max 20) points, LLM-judged rubric score. Each harness remains a separate evaluation population. Score maximum comes from the current source header; no fixed old project-count bound. Source exact model/effort labels and consistency penalty retained.
Metric: Python Code Quality (max 20) points, LLM-judged rubric score. Source: aicodingdaily.com. Saturation forecast: Around February 2027. 23 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | DeepSeek V4 Pro (0813) (Max) | 17.83 | #81 (DeepSeek V4 Pro (0813)) |
| 2 | Kimi K3.1 Pro (High) | 17.8 | #28 (Kimi K3.1 Pro) |
| 3 | DeepSeek V4 Pro (0813) (High) | 17.3 | #81 (DeepSeek V4 Pro (0813)) |
| 4 | GLM-5.3 Flash (Max) | 17 | #119 (GLM-5.3 Flash) |
| 5 | MiniMax-M3.1-Flash (Max) | 16.9 | #104 (MiniMax-M3.1-Flash) |
| 6 | DeepSeek V4.1 Flash (1017) (High) | 16.7 | #38 (DeepSeek V4.1 Flash (1017)) |
| 7 | Kimi K3.1 Pro (Max) | 16.47 | #28 (Kimi K3.1 Pro) |
| 8 | Qwen 3.8 Flash (xHigh) | 16.33 | #109 (Qwen 3.8 Flash) |
| 9 | DeepSeek V4.1 Flash (1017) (Max) | 16.27 | #38 (DeepSeek V4.1 Flash (1017)) |
| 10 | GLM-5.3 (High) | 16.27 | #62 (GLM-5.3) |
| 11 | Qwen 3.8 27B (xHigh) | 15.3 | #199 (Qwen 3.8 27B) |
| 12 | Mistral Large 4 (High) | 15.03 | #125 (Mistral Large 4) |
| 13 | Hy3 (High) | 14.03 | #193 (Hy3) |
| 14 | Mercury 3 (High) | 12.73 | #187 (Mercury 3) |
| 15 | Muse Spark 1.3 (xHigh) | 11.9 | #42 (Muse Spark 1.3) |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=ai-coding-daily-opencode-python-code-quality · How It Works · Data refreshed daily, snapshot 2026-11-12.