Transform your 3D texturing workflow with the power of generative AI, directly within Blender!
update readme, manual installation, bump version, make uv unwrapping default for baking
sakalond1fe93e408d9a
diff --git a/README.md b/README.mdindex c9cd28d..a476d0e 100644--- a/README.md+++ b/README.md@@ -44,7 +44,7 @@ StableGen brings AI-powered 3D generation and texturing directly into Blender:* **Flexible texture pipeline:** Use TRELLIS.2's native PBR textures, or automatically texture the generated mesh with SDXL, FLUX.1-dev, or Qwen Image Edit for higher-quality diffusion textures.* **Preview Gallery:** Generate multiple candidate images with different seeds and pick the best before committing to 3D generation.* **Smart mesh handling:** Auto-recovery from mesh corruption, configurable decimation/remeshing, import scaling, and studio lighting setup.- * VRAM-conscious: disk offloading, configurable attention backend, and low-VRAM modes.+ * VRAM-conscious: disk offloading, configurable attention backend* Powered by [ComfyUI-TRELLIS2](https://github.com/PozzettiAndrea/ComfyUI-TRELLIS2) (installable via `installer.py`).* 🌍 **Scene-Wide Multi-Mesh Texturing:*** Don't just texture one mesh at a time! StableGen is designed to apply textures to **all mesh objects in your scene simultaneously** from your defined camera viewpoints. Alternatively, you can choose to texture only selected objects.@@ -94,7 +94,8 @@ StableGen brings AI-powered 3D generation and texturing directly into Blender:## 🚀 Showcase Gallery-See what StableGen can do!+<details open>+<summary>See what StableGen can do!</summary><sub>Tip: Refresh the page to synchronize all GIF animations.</sub>@@ -272,6 +273,8 @@ Texturing a complex scene consisting of many mesh objects.</details>+</details>+---## 🛠️ How It Works (A Glimpse)@@ -568,7 +571,7 @@ The TRELLIS.2 feature relies on several third-party components, each with its ow**All other TRELLIS.2 modes do not introduce licensing restrictions:*** **Shape-only mode ("None")** - does not use nvdiffrast/nvdiffrec. All other pipeline components are permissively licensed (MIT/Apache 2.0 + DINOv3 License).-* **Projection-based texture modes ("SDXL", "Flux 1", "Qwen Image Edit")** - does not use nvdiffrast/nvdiffrec. The licensing terms of the selected diffusion model apply as usual (e.g., FLUX.1-dev has its own license terms separate from the TRELLIS.2 pipeline).+* **Projection-based texture modes ("SDXL", "Qwen Image Edit", ...)** - do not use nvdiffrast/nvdiffrec. The licensing terms of the selected diffusion model apply as usual (e.g., FLUX.1-dev has its own license terms separate from the TRELLIS.2 pipeline).If you require commercial use of the "Native (TRELLIS.2)" texture mode, consider contacting NVIDIA regarding commercial licensing for nvdiffrast/nvdiffrec.@@ -629,4 +632,4 @@ Ondřej Sakala* X/Twitter: `@sakalond`----*Last Updated: February 14, 2026*+*Last Updated: March 5, 2026*diff --git a/docs/MANUAL_INSTALLATION.md b/docs/MANUAL_INSTALLATION.mdindex af51e1a..d711ee3 100644--- a/docs/MANUAL_INSTALLATION.md+++ b/docs/MANUAL_INSTALLATION.md@@ -270,3 +270,124 @@ For users with NVIDIA GPUs on Linux or Windows (WSL2 recommended for best perfor* **License:** Apache 2.0* **Size:** ~12.7 GB+**8. (Optional) TRELLIS.2 - Image/Text to 3D Setup**++Generate 3D meshes from a single image or text prompt. Models auto-download on first use (~2 GB each for shape and texture stages).++> **Note:** The TRELLIS.2 native textured output pipeline uses NVIDIA non-commercial libraries (nvdiffrast). Shape-only and SDXL/Qwen texture generation does not have this restriction.++* **a) Install the ComfyUI-TRELLIS2 Custom Node**+ 1. Navigate to your ComfyUI custom nodes directory:+ ```bash+ cd <YourComfyUIDirectory>/custom_nodes/+ ```+ 2. Clone the repository at the tested commit:+ ```bash+ git clone https://github.com/PozzettiAndrea/ComfyUI-TRELLIS2.git+ cd ComfyUI-TRELLIS2+ git checkout 6b0b2148f45bbafa0b86bfd25c63602b63a7aae0+ ```+ 3. Install its dependencies (from inside the cloned folder):+ ```bash+ pip install -r requirements.txt+ ```+ 4. Install the comfy-env dependency:+ ```bash+ pip install comfy-env==0.2.0+ ```+ 5. **Restart ComfyUI.**+ * **License:** MIT (node code)+ * **Size:** ~0.1 GB (models auto-download on first use)++* **b) Important: Apply Patches (Recommended)**++ The `installer.py` script automatically applies several critical patches to ComfyUI-TRELLIS2 that fix major VRAM leaks (up to ~16 GB savings) and improve stability. If you install manually, it is **strongly recommended** to run the installer for this step, or apply the patches yourself. Key patches include:+ * Wrapping DinoV3 in `torch.no_grad()` to prevent 5-9 GB GPU memory leak.+ * Unregistering models from comfy-env's registry to prevent ~7 GB VRAM leak.+ * Cleaning up temporary IPC files between generations.+ * Python 3.10 compatibility fix for `Path.is_junction`.++**9. (Optional) PBR Decomposition - Marigold IID Setup**++Decompose color textures into PBR material maps (roughness, metallic, normal, height, albedo). Models auto-download on first use.++* **a) Install the ComfyUI-Marigold Custom Node**+ 1. Navigate to your ComfyUI custom nodes directory:+ ```bash+ cd <YourComfyUIDirectory>/custom_nodes/+ ```+ 2. Clone the repository:+ ```bash+ git clone https://github.com/kijai/ComfyUI-Marigold.git+ ```+ 3. Install its dependencies:+ ```bash+ cd ComfyUI-Marigold+ pip install -r requirements.txt+ pip install "diffusers>=0.28"+ ```+ 4. **(Windows only)** Install triton:+ ```bash+ pip install triton-windows+ ```+ 5. **Restart ComfyUI.**+ * **License:** GPL-3.0+ * **Size:** ~0.01 GB (models auto-download on first use, ~2 GB each)++**10. (Optional) PBR Decomposition - StableDelight Setup**++Adds a specular-free albedo extraction option for PBR decomposition. **Requires Marigold IID (step 9) to also be installed.**++* **a) Install the ComfyUI_StableDelight_ll Custom Node**+ 1. Navigate to your ComfyUI custom nodes directory:+ ```bash+ cd <YourComfyUIDirectory>/custom_nodes/+ ```+ 2. Clone the repository:+ ```bash+ git clone https://github.com/lldacing/ComfyUI_StableDelight_ll.git+ ```+ 3. **Restart ComfyUI.**+ * **License:** Apache 2.0++* **b) Download the StableDelight Model**+ * **Directory:** `<YourComfyUIDirectory>/models/diffusers/Stable-X--yoso-delight-v0-4-base`+ * **Source:** [https://huggingface.co/Stable-X/yoso-delight-v0-4-base](https://huggingface.co/Stable-X/yoso-delight-v0-4-base)+ * You can download the full repo or use:+ ```bash+ pip install huggingface_hub+ python -c "from huggingface_hub import snapshot_download; snapshot_download('Stable-X/yoso-delight-v0-4-base', local_dir='<YourComfyUIDirectory>/models/diffusers/Stable-X--yoso-delight-v0-4-base')"+ ```+ * **License:** Apache 2.0+ * **Size:** ~3.3 GB++**11. (Optional) FLUX.2 Klein Setup (Experimental)**++> **⚠️ Experimental:** Klein is not yet fully stable - there are known issues with geometry guidance. Expect improvements in future updates.++FLUX.2 Klein is a multi-reference image editing architecture from Black Forest Labs. No custom nodes required - all nodes are built into ComfyUI.++* **a) Download the Klein 4B Diffusion Model (FP8)**+ * **Directory:** `<YourComfyUIDirectory>/models/diffusion_models/`+ * **Filename:** `flux-2-klein-base-4b-fp8.safetensors`+ * **Download URL:** [https://huggingface.co/black-forest-labs/FLUX.2-klein-base-4b-fp8/resolve/main/flux-2-klein-base-4b-fp8.safetensors?download=true](https://huggingface.co/black-forest-labs/FLUX.2-klein-base-4b-fp8/resolve/main/flux-2-klein-base-4b-fp8.safetensors?download=true)+ * **License:** Apache 2.0+ * **Size:** ~4.1 GB++* **b) Download the Qwen 3 4B Text Encoder (bf16)**+ * **Directory:** `<YourComfyUIDirectory>/models/text_encoders/`+ * **Filename:** `qwen_3_4b.safetensors`+ * **Download URL:** [https://huggingface.co/Comfy-Org/z_image_turbo/resolve/main/split_files/text_encoders/qwen_3_4b.safetensors?download=true](https://huggingface.co/Comfy-Org/z_image_turbo/resolve/main/split_files/text_encoders/qwen_3_4b.safetensors?download=true)+ * **License:** Apache 2.0+ * **Size:** ~8.1 GB++* **c) Download the FLUX.2 VAE**+ * **Directory:** `<YourComfyUIDirectory>/models/vae/`+ * **Filename:** `flux2-vae.safetensors`+ * **Download URL:** [https://huggingface.co/Comfy-Org/flux2-dev/resolve/main/split_files/vae/flux2-vae.safetensors?download=true](https://huggingface.co/Comfy-Org/flux2-dev/resolve/main/split_files/vae/flux2-vae.safetensors?download=true)+ * **License:** Apache 2.0+ * **Size:** ~321 MB++* **Total FLUX.2 Klein download size:** ~12.4 GB+* **Estimated VRAM requirement:** ~13 GB+diff --git a/stablegen/__init__.py b/stablegen/__init__.pyindex f4eb45b..f132985 100644--- a/stablegen/__init__.py+++ b/stablegen/__init__.py@@ -16,7 +16,7 @@ bl_info = {"name": "StableGen","category": "Object","author": "Ondrej Sakala",- "version": (0, 2, 0),+ "version": (0, 3, 0),'blender': (4, 2, 0)}diff --git a/stablegen/blender_manifest.toml b/stablegen/blender_manifest.tomlindex 30f36eb..6d80c93 100644--- a/stablegen/blender_manifest.toml+++ b/stablegen/blender_manifest.toml@@ -1,15 +1,15 @@schema_version = "1.0.0"id = "stablegen"-version = "0.2.0"+version = "0.3.0"name = "StableGen"-tagline = "AI-Powered Texturing"+tagline = "AI-Powered 3D Asset Generation & Texturing"maintainer = "Ondrej Sakala <[email protected]>"type = "add-on"blender_version_min = "4.2.0"website = "https://github.com/sakalond/StableGen"-tags = ["Material", "Paint", "UV"]+tags = ["Material", "Paint", "UV", "Mesh", "Import-Export"]license = ["SPDX:GPL-3.0-or-later"]@@ -43,5 +43,5 @@ wheels = [][permissions]-network = "Connects to a local ComfyUI server for AI texture generation"-files = "Saves generated textures, intermediate files, exported GIFs"+network = "Connects to a local ComfyUI server for AI generation"+files = "Saves generated textures, 3D meshes, PBR maps, and exported GIFs"diff --git a/stablegen/render_tools.py b/stablegen/render_tools.pyindex b92476f..937f412 100644--- a/stablegen/render_tools.py+++ b/stablegen/render_tools.py@@ -5119,7 +5119,7 @@ class BakeTextures(bpy.types.Operator):('lightmap', 'Lightmap Pack', 'Use Lightmap Pack with default parameters'),('pack', 'Pack Islands', 'Use Pack Islands with default parameters')],- default='none'+ default='smart') # type: ignoreadd_material: bpy.props.BoolProperty(