Aider polyglot coding leaderboard
Aider polyglot coding leaderboard is run by Aider. It has ranked 69 models, of which the catalogue holds 40, scoring from 3.6 to 88 on Percent correct.
measured by Aider · the board itself
| Place | Model | Metric | Score |
|---|---|---|---|
| 1stof 69 | GPT-5 | Percent correct (high) | 88 |
| 3rdof 69 | o3-pro | Percent correct (high) | 84.9 |
| 4thof 69 | Gemini 2.5 Pro Preview 06-05 | Percent correct | 83.1 |
| 6thof 69 | o3 | Percent correct (high) | 81.3 |
| 7thof 69 | Grok 4 | Percent correct (high) | 79.6 |
| 9thof 69 | GPT-4.1 | Percent correct | 78.2 |
| 11thof 69 | Gemini 2.5 Pro Preview 05-06 | Percent correct | 76.9 |
| 12thof 69 | DeepSeek V3.2 Exp | Percent correct | 74.2 |
| 13thof 69 | Gemini 2.5 Pro | Percent correct | 72.9 |
| 14thof 69 | Claude Opus 4 | Percent correct | 72 |
| 15thof 69 | o4 Mini High | Percent correct (high) | 72 |
| 16thof 69 | DeepSeek-R1 (0528) | Percent correct | 71.4 |
| 19thof 69 | Claude 3.7 Sonnet | Percent correct | 64.9 |
| 20thof 69 | Claude 3.5 Sonnet | Percent correct | 64 |
| 21stof 69 | o1 | Percent correct (high) | 61.7 |
| 22ndof 69 | Claude Sonnet 4 | Percent correct | 61.3 |
| 24thof 69 | o3 Mini High | Percent correct (high) | 60.4 |
| 25thof 69 | Qwen3 235B A22B | Percent correct | 59.6 |
| 26thof 69 | Kimi K2 0711 | Percent correct | 59.1 |
| 27thof 69 | DeepSeek-R1 | Percent correct | 56.9 |
| 29thof 69 | Gemini 2.5 Flash | Percent correct | 55.1 |
| 30thof 69 | DeepSeek-V3 0324 | Percent correct | 55.1 |
| 32ndof 69 | o3-mini | Percent correct (medium) | 53.8 |
| 40thof 69 | OpenAI ChatGPT-4o | Percent correct | 45.3 |
| 43rdof 69 | GPT OSS 120B | Percent correct (high) | 41.8 |
| 44thof 69 | Qwen3 32B | Percent correct | 40 |
| 48thof 69 | o1-mini | Percent correct | 32.9 |
| 49thof 69 | GPT-4.1 mini | Percent correct | 32.4 |
| 50thof 69 | Claude Haiku 3.5 | Percent correct | 28 |
| 53rdof 69 | GPT-4o (2024-08-06) | Percent correct | 23.1 |
| 54thof 69 | Gemini 2.0 Flash | Percent correct | 22.2 |
| 56thof 69 | QwQ-32B | Percent correct | 20.9 |
| 58thof 69 | GPT-4o (2024-11-20) | Percent correct | 18.2 |
| 60thof 69 | Qwen2.5 Coder 32B Instruct | Percent correct | 16.4 |
| 61stof 69 | Llama 4 Maverick 17B | Percent correct | 15.6 |
| 64thof 69 | Codestral 25.01 | Percent correct | 11.1 |
| 65thof 69 | openhands-lm-32b-v0.1 | Percent correct | 10.2 |
| 66thof 69 | GPT-4.1 nano | Percent correct | 8.9 |
| 68thof 69 | Gemma 3 27B | Percent correct | 4.9 |
| 69thof 69 | GPT-4o-mini (2024-07-18) | Percent correct | 3.6 |
Read from the board on 2026-09-16