LLM2014 Code 2025-07: leaderboard
LLM2014 monthly coding benchmark for 2025-07, reporting the aggregate code-generation score across programming tasks.
Metric: Median Score. Source: raw.githubusercontent.com. Saturation forecast: Estimated already saturated. 19 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Sonnet 4 (Thinking) | 66.98 | #199 (Claude Sonnet 4) |
| 2 | O4 Mini (High) | 66.75 | #180 (O4 Mini) |
| 3 | Claude Sonnet 4 | 61.02 | #199 |
| 4 | Gemini 2.5 Pro | 58.05 | #152 |
| 5 | Grok 4 | 55.12 | #181 |
| 6 | GPT-4.1 Mini | 50.87 | #355 |
| 7 | Qwen 3 235B A22B (Thinking) | 49.83 | #313 (Qwen 3 235B A22B) |
| 8 | DeepSeek R1 0528 | 49.25 | #222 |
| 9 | Gemini 2.5 Flash (Thinking) | 45.04 | #245 (Gemini 2.5 Flash) |
| 10 | Seed-1.6 (Thinking) | 43.76 | #265 (Seed-1.6) |
| 11 | Grok 3 Mini | 39.58 | #281 |
| 12 | DeepSeek V3 (0324) | 38.04 | #340 |
| 13 | Seed-1.6 | 34.99 | #265 |
| 14 | Qwen 3 235B A22B | 34.96 | #313 |
| 15 | Kimi K2 (0711) | 27.14 | #492 |
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-2025-07 · How It Works · Data refreshed daily, snapshot 2026-10-20.