P3D-Bench - Text-to-3D Parametric Geometry: leaderboard
P3D-Bench (2026) Text-to-3D from parametric specifications (explicit dimensions and placements) on 400 cases: geometric fidelity of the generated JSON or OpenSCAD model to the ground truth (F-scores, normal consistency, Chamfer distance and IoU). Programs are compiled, meshed and aligned to the ground truth, invalid outputs take the worst value, and each general model runs with its maximum thinking budget; the score is on a 0-100 scale averaged over output formats, higher is better.
Metric: Geometry bucket score (0-100) (equal-weight mean of normalized F-scores at 0.05 and 0.01, normal consistency, Chamfer distance and IoU) on 400 Text-to-3D cases from parametric specifications (minimal JSON and OpenSCAD outputs), the model writes a parametric CAD program that is compiled, meshed and aligned to the ground truth; invalid outputs take the worst value; each general model runs with its maximum thinking budget; averaged over the output formats; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 9 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.5 (xHigh) | 70.6 | #49 (GPT-5.5) |
| 2 | Gemini 3.1 Pro (Preview) (High) | 69.9 | #94 (Gemini 3.1 Pro (Preview)) |
| 3 | Claude Opus 4.6 (Max) | 69.2 | #102 (Claude Opus 4.6) |
| 4 | Qwen 3.6 Plus (Thinking) | 65 | #222 (Qwen 3.6 Plus) |
| 5 | DeepSeek V4 Pro (Max) | 64.7 | #188 (DeepSeek V4 Pro) |
| 6 | Seed 2.0 Pro (High) | 63.3 | #211 (Seed 2.0 Pro) |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=p3d-bench-text-to-3d-parametric-geometry · How It Works · Data refreshed daily, snapshot 2026-11-13.