CodeARC - Annotated: leaderboard
Metric: Success Rate (%; descriptive function names, 1,114 functions). Source: arxiv.org. Saturation forecast: Estimated already saturated. 18 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | O3 Mini | 59.5 |
| 2 | O1 Mini | 53.2 |
| 3 | DeepSeek R1 | 49.8 |
| 4 | Claude 3.7 Sonnet | 39 |
| 5 | Llama 3.1 405B Instruct | 38.6 |
| 6 | GPT-4o | 37.8 |
| 7 | DeepSeek V3 | 37.7 |
| 8 | Llama 3.1 70B Instruct | 30.1 |
| 9 | Qwen 2.5 72B Instruct | 30.1 |
| 10 | Qwen 2.5 7B Instruct | 29.2 |
| 11 | Llama 3.2 90B Vision Instruct | 28.4 |
| 12 | GPT-4o Mini | 26.1 |
| 13 | Mixtral 8x22B Instruct (v0.1) | 25.1 |
| 14 | Llama 3.2 11B Instruct | 24.9 |
| 15 | Mixtral 8x7B Instruct (v0.1) | 20.3 |
Interactive version: theaggregate.ai/benchmark?slug=codearc-annotated · How It Works · Data refreshed daily, snapshot 2026-09-24.