LH-Bench (Figma-to-Code) - Skill Score: leaderboard
Metric: Process-tier skill score (1-5): the mean of three LLM judges (Gemini 3.1 Pro, Claude Sonnet 4.6, GPT-5.2) scoring the run's structured traces on four expert-authored process rubrics (design inspection, token and style extraction, component architecture, build verification), on the 33 LH-Bench Figma-to-code tasks (rebuild a real Figma design as a working front end through the Figma API over MCP, following an expert-authored SKILL.md workflow), the model running autonomously in its vendor's agent harness; higher is better. Source: arxiv.org. Saturation forecast: Around October 2027. 3 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 4.6 | 3.27 | #60 |
| 2 | GPT-5.2 | 3.16 | #105 |
| 3 | Gemini 3.1 Pro (Preview) | 2.8 | #54 |
No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #8 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=lh-bench-figma-to-code-skill-score · How It Works · Data refreshed daily, snapshot 2026-10-11.