LLM2014 Code 2025-11 - Python: leaderboard
LLM2014 monthly coding benchmark for 2025-11, measuring Python code-generation problem-solving on the source score scale.
Metric: Score. Source: raw.githubusercontent.com. Saturation forecast: Estimated already saturated. 26 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Pro | 10 | #80 |
| 2 | Claude Opus 4.5 | 9.33 | #83 |
| 3 | GPT-5 Mini (High) | 9.2 | #186 (GPT-5 Mini) |
| 4 | Claude Sonnet 4.5 (Thinking) | 8.88 | #142 (Claude Sonnet 4.5) |
| 5 | DeepSeek V3.2 (Thinking) | 8.08 | #205 (DeepSeek V3.2) |
| 6 | Gemini 2.5 Pro | 7.9 | #152 |
| 7 | Qwen 3 Max | 7.62 | #208 |
| 8 | MiniMax-M2 | 7.57 | #317 |
| 9 | GLM-4.6 (Thinking) | 7.22 | #252 (GLM-4.6) |
| 10 | Claude Sonnet 4.5 | 7.15 | #142 |
| 11 | GLM-4.6 | 6.75 | #252 |
| 12 | DeepSeek V3.2 | 6.72 | #205 |
| 13 | Kimi K2 (Thinking) | 6.42 | #243 (Kimi K2) |
| 14 | Ling-1T | 5.4 | #348 |
| 15 | Seed-1.6 (Thinking) | 5.4 | #265 (Seed-1.6) |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=llm2014-code-2025-11-python · How It Works · Data refreshed daily, snapshot 2026-10-20.