Spaces:
Paused
Paused
| # Model decisions | |
| Checked September 16–17, 2026. Model and source revisions are in `models.lock.json`. | |
| These are implementation choices for ZeroGPU, not results of a comparative benchmark. | |
| | Stage | Choice | Reason and tradeoff | | |
| | --- | --- | --- | | |
| | Prompt/image conditioning | [FLUX.2 Klein 4B](https://huggingface.co/black-forest-labs/FLUX.2-klein-4B) | Current compact unified generation/editing model, four distilled steps, about 13GB VRAM. Chosen over heavier Qwen-Image and FLUX.2 variants for lease time and memory. | | |
| | Textured 3D | [TRELLIS.2 4B](https://huggingface.co/microsoft/TRELLIS.2-4B) | Current Microsoft structured-latent geometry and PBR model with an official [ZeroGPU implementation](https://huggingface.co/spaces/microsoft/TRELLIS.2). Use low texture settings here because SR is a separate stage. | | |
| | Part decomposition | [NVIDIA PartField](https://github.com/nv-tlabs/PartField) | Feedforward learned part features across categories, followed by feature clustering. Uses the released Objaverse checkpoint. Samples original faces without remeshing so UV correspondence survives. | | |
| | Texture upscale | [Real-ESRGAN x4plus](https://github.com/xinntao/Real-ESRGAN) | Deliberately older, conservative SR baseline. Newer SeedVR2 diffusion restoration can invent texture details and distort UV islands; it is heavier and optimized for images/video. Only base color uses neural SR. A measured bake comparison is needed before claiming either wins on this workload. | | |
| | Refinement | `inclusionai/ling-3.0-flash-vl:free` | Exact user-selected model through OpenRouter and [fx](https://github.com/vercel-labs/fx). Availability of the temporary free route is not guaranteed. | | |
| [TheOrcDev game-model-cleanup](https://github.com/TheOrcDev/skills/tree/main/game-dev/game-model-cleanup) | |
| and its workflow are loaded verbatim into the fx task, with an explicit mapping from its | |
| Final Stand project paths/contracts to this run's immutable source, component inventory, | |
| rendered evidence and repair schema. Other skills in that repository concern animation, | |
| Unity integration or unrelated application work and are not silently applied. | |
| PartField's single encoder `torch_scatter.scatter_mean` import is adapted to native | |
| PyTorch `scatter_add_` plus counts. This avoids compiling a version-specific scatter | |
| extension and leaves checkpoint architecture/weights unchanged. The adapter does not | |
| use the upstream Lightning trainer, which would start distributed workers inside a lease. | |
| Deployment follows [Hugging Face ZeroGPU](https://huggingface.co/docs/hub/spaces-zerogpu): | |
| Gradio SDK, startup model loading, bounded GPU functions, no torch.compile. TRELLIS retains | |
| its upstream low-memory model staging; its transfers can increase lease time. The 10% | |
| reserve is admission control against measured live availability, not a promise that CUDA | |
| fragmentation or external allocations can never cause OOM. | |