VibeCodingBench: leaderboard
Production-oriented coding benchmark evaluating AI coding agents across functional correctness, visual fidelity, code quality, security, cost, and speed on representative developer tasks.
Metric: Average Score. Source: github.com. 15 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 4.5 | 89.15 | #83 |
| 2 | Claude Haiku 4.5 | 88.97 | #280 |
| 3 | Grok 4 Fast | 88.8 | #254 |
| 4 | OpenAI GPT-5.2 | 88.75 | |
| 5 | Qwen3 Max | 88.6 | |
| 6 | Claude Sonnet 4.5 | 88.56 | #142 |
| 7 | GLM 4-Plus | 88.2 | |
| 8 | DeepSeek v3.2 | 88.19 | |
| 9 | Grok 4 | 88 | #181 |
| 10 | MiniMax M2.1 | 87.42 |
Interactive version: theaggregate.ai/benchmark?slug=vibecodingbench · How It Works · Data refreshed daily, snapshot 2026-10-20.