Aider polyglot coding leaderboard: leaderboard
225 Exercism coding challenges across C++, Go, Java, JavaScript, Python, and Rust, with two attempts per problem to test code generation and debugging.
Metric: Pass rate (%). Source: aider.chat. Source: benchmarklist.com. Saturation forecast: Estimated already saturated. 69 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5 (High) | 88 | #98 (GPT-5) |
| 2 | GPT-5 (Medium) | 86.7 | #98 (GPT-5) |
| 3 | O3 (High) | 81.3 | #124 (O3) |
| 4 | GPT-5 (Low) | 81.3 | #98 (GPT-5) |
| 5 | O3 | 76.9 | #124 |
| 6 | Gemini 2.5 Pro (Preview 05-06) | 76.9 | #147 |
| 7 | DeepSeek V3.2 Exp | 74.2 | #234 |
| 8 | Gemini 2.5 Pro (Preview 03-25) | 72.9 | #123 |
| 9 | O4 Mini (High) | 72 | #180 (O4 Mini) |
| 10 | Claude Opus 4 (20250514) (Thinking) | 72 | #154 (Claude Opus 4 (20250514)) |
| 11 | DeepSeek R1 0528 | 71.4 | #222 |
| 12 | Claude Opus 4 (20250514) | 70.7 | #154 |
| 13 | Claude Opus 4 (Non-reasoning) | 70.7 | #163 (Claude Opus 4) |
| 14 | Claude 3.7 Sonnet (20250219) (Thinking) | 64.9 | #201 (Claude 3.7 Sonnet (20250219)) |
| 15 | O1 (2024-12-17) (High) | 61.7 | #149 (O1 (2024-12-17)) |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=aider-polyglot-coding-leaderboard · How It Works · Data refreshed daily, snapshot 2026-10-20.