CCBench — leaderboard
CCBench evaluates model capability on coding & software engineering tasks from the linked upstream source with Success Rate as the primary reported metric.
Metric: Accuracy (%). Source: huggingface.co. Status: saturation imminent. 229 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | InternVL3-78B | 83.5 |
| 2 | InternVL3-38B | 80.6 |
| 3 | InternVL3-8B | 77.8 |
| 4 | InternVL2-8B | 77.1 |
| 5 | InternVL3-14B | 77.1 |
| 6 | GPT-4.1 (2025-04-14) | 73.5 |
| 7 | Qwen 2 VL 72B | 69.8 |
| 8 | Qwen 2 VL 7B | 65.7 |
| 9 | GPT-4.1 Mini | 64.3 |
| 10 | Ovis2-8B | 59.6 |
| 11 | Grok 2 (1212) | 57.3 |
| 12 | Claude 3.5 Sonnet (20241022) | 56.1 |
| 13 | MiniCPM-V-2.6 | 55.7 |
| 14 | Claude 3.7 Sonnet | 55.1 |
| 15 | Claude 3.5 Sonnet | 54.1 |
Interactive version: theaggregate.ai/benchmark?slug=ccbench · How the rankings work · Data refreshed daily, snapshot 2026-07-22.