Gemini plus ChatGPT and Blender MCP can now build 3D models from one photo
FinanceYF5 · x · 2026-07-27
A creator shows a workflow for generating a 3D model by chaining Gemini, Blender with MCP, Tencent Hunyuan 3D, and ChatGPT.
- Step 1: use Gemini to generate a clean front reference image from a model photo.
- Step 2: feed that front view into ChatGPT to synthesize additional angles, such as a back view and a 45° front-left view.
- The post frames this as a practical way AI can help build almost any 3D asset, and the attached image shows the resulting model-like bust.
More from coding & agent
- He benchmarks new models by having them build their own copilot — $28 of compute — bucolucas · 2026-10-27
- Founder generates 46 app promo videos in 4 days with Claude, skipping $11,500 in UGC fees — PrajwalTomar_ · 2026-10-27
- Should agents be allowed to change plans mid-task? Reddit debates the drift boundary — wistfulcracker84 · 2026-10-27
- HeyGen + Canva MCP turns slide speaker notes into an avatar presentation video — HeyGen · 2026-10-27
- Local beats cloud: Qwen 3.8 FN handles private student data as well as frontier models — dr_lm · 2026-10-27
- Cloudflare Computer adds agent interruption recovery and approval pauses in v0.7.0 — irvinebroque · 2026-10-27