Vision2Web S3-2026 (Claude Code) - Level 3 Full-Stack: leaderboard
Vision2Web Level 3: the agent develops a full-stack web application from requirements and prototypes; the score is the mean of a VLM-judged visual score and a functional score from workflow-driven GUI tests, on a 0-100 scale under the Claude Code framework, higher is better (He et al., 2026). Season S3-2026 (Jul to Sep 2026) uses Gemini-3.5-flash as the visual judge and GLM-5V-Turbo (Claude Code) as the GUI testing agent; scores compare only within a season.
Metric: Score (0-100). Source: vision2web-bench.github.io. Saturation forecast: Around February 2027. 6 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 4.8 | 69.8 | #56 |
| 2 | Claude Opus 4.7 | 63.9 | #75 |
| 3 | Claude Sonnet 4.6 | 58 | #158 |
| 4 | MiniMax-M3 | 57.4 | #265 |
| 5 | Claude Opus 4.6 | 56.5 | #100 |
| 6 | Kimi K2.6 | 50.2 | #173 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=vision2web-s3-2026-claude-code-level-3-full-stack · How It Works · Data refreshed daily, snapshot 2026-11-12.