OpenAI GPT-6 Astra creates textured 3D objects or scenes from an image or prompt with material-aware PBR textures for timber, stone, glass, metal, fabric and ceramic. Each generation costs $8 and can take a long time; results are interpretations rather than dimensionally exact scans.
Added Sep 9, 2026
Price per model
$8.00/model
Model Type
both
Typical Duration
~15 min
Inputs
Image + Text
Price table
Each row is the price for one completed 3D model generation.
| Variant | Price |
|---|---|
Default Default | $8.00/model |
Settings
Generation controls available for this model.
Output Format
GLB
Typical Duration
~15 min
Subject size (m)
Default
1
The longest subject dimension in meters.
API request
Submit the generation with POST /api/generate-video, then poll the returned run ID for completion.
{
"model": "openai/gpt-6-astra",
"prompt": "A studio-quality 3D model of a ceramic desk lamp.",
"size": 1
}Browse the 3D models API catalogRelated 3d models
Compare GPT-6 Astra with similar models from the same provider or model family.
Seed3D 2.0
bytedance/seed3d-2.0ByteDance image-to-3D with an optional guiding prompt. Produces detailed GLB assets; generations typically take several minutes.
Rodin 2.5
hyper3d/rodin-v2.5Production-ready 3D assets from an image or prompt with clean topology, UVs and PBR textures. Quality tiers and a high-detail add-on change the price.
Tripo 2.5
tripo3d/v2.5Tripo image-to-3D. Converts a single source image into a textured GLB in about a minute at a low fixed price.
Hunyuan 3D 3.1 Rapid
wavespeed-ai/hunyuan-3d-v3.1-rapidFast, low-cost draft mesh from a single image or a short prompt (up to 200 characters). Geometry only, no textures, in under a minute. Use Hunyuan3D V3 or Tripo for a textured result.
Hunyuan3D V3
wavespeed-ai/hunyuan3d-v3Best-quality textured 3D from an image or prompt: film-quality geometry with baked color textures, optional PBR materials and low-poly output. About two minutes; files can reach 30 MB.
Meshy 6
wavespeed-ai/meshy6Meshy 6 turns an image or prompt (up to 600 characters) into a textured, remeshed 3D model with controllable topology and polycount. Generations typically take several minutes.