LIVE
Anthropic files IPO prospectus, with risk disclosures taking center stage26/10/26 · Anthropic|Robot policy trained only in simulation matches top real-world systems26/10/26|OpenAI unveils GPT-6.1 Terra, a model built for computer use26/10/26 · OpenAI|OpenAI models are coming to Google Cloud21/10/26 · OpenAI|OpenAI details ChatGPT's new advertising infrastructure21/10/26 · OpenAI|Liquid AI unveils LFM3, a base for specialized multimodal models27/10/26 · Liquid AI|US Police Departments Use AI to Flag 'Suspicious' Driving Behavior27/10/26|AllenAI questions whether LLMs can judge research novelty27/10/26 · AllenAI|Open-Source AI Tools Aim to Speed Up Dementia Research27/10/26 · NVIDIA|Solid-state transformers eyed as a fix for AI's power grid bottleneck27/10/26|OpenAI expands Oracle partnership to bring business apps into ChatGPT27/10/26 · OpenAI|OpenAI publishes a page announcing 'GPT-6.1 Terra', with no technical details available26/10/26 · OpenAI|Anthropic files IPO prospectus, with risk disclosures taking center stage26/10/26 · Anthropic|Robot policy trained only in simulation matches top real-world systems26/10/26|OpenAI unveils GPT-6.1 Terra, a model built for computer use26/10/26 · OpenAI|OpenAI models are coming to Google Cloud21/10/26 · OpenAI|OpenAI details ChatGPT's new advertising infrastructure21/10/26 · OpenAI|Liquid AI unveils LFM3, a base for specialized multimodal models27/10/26 · Liquid AI|US Police Departments Use AI to Flag 'Suspicious' Driving Behavior27/10/26|AllenAI questions whether LLMs can judge research novelty27/10/26 · AllenAI|Open-Source AI Tools Aim to Speed Up Dementia Research27/10/26 · NVIDIA|Solid-state transformers eyed as a fix for AI's power grid bottleneck27/10/26|OpenAI expands Oracle partnership to bring business apps into ChatGPT27/10/26 · OpenAI|OpenAI publishes a page announcing 'GPT-6.1 Terra', with no technical details available26/10/26 · OpenAI|
AppWatch outofficial publisher

Hunyuan3D 2.1 (Tencent)

App : A product you use directly

Textured PBR 3D generation from an image, running locally.

The best open PBR texture generator, but the licence is not applicable in the EU: avoid for commercial use here.

Overall score
6.6 /10
Security: one caveat, see the card
Utility
8.5
Reliability
4.0
Security
One caveat
Ease
Technical · Technical

Security reflects where data goes, requested permissions and incidents on record at the verification date. It is not a guarantee of zero risk: check the deployment mode and the connections you enable.

What it does, concretely

Creating a quality textured 3D object from a single image normally requires hours of manual work in specialised software. Hunyuan3D 2.1 automates this process by first generating the geometry, then applying high-resolution PBR (Physically Based Rendering) textures, all running locally on your machine.

The system works in two distinct stages: a diffusion model (Hunyuan3D-DiT, 3 billion parameters) generates the 3D mesh from a reference image, then a second model (Hunyuan3D-Paint, 1.3 billion parameters) applies PBR textures to that mesh. Both models have been fully open-sourced since June 2025, including training code.

This tool targets game developers, 3D artists and technically minded creators who are comfortable installing a Python environment and managing software dependencies. A Gradio interface is available to avoid writing code, and a Blender add-on allows generation to be integrated directly into that software.

However, Tencent's community licence is not compatible with commercial use in the European Union, ruling out any professional exploitation for Europe-based users. People without command-line experience or with less than 16 GB of VRAM for full generation (shape and texture) will face significant hurdles.

What it offers

  • ·3D mesh generation from a single image using the Hunyuan3D-DiT model (3 billion parameters).
  • ·High-resolution PBR texture application on generated or hand-crafted meshes via Hunyuan3D-Paint.
  • ·Lightweight models available: mini version (0.6 billion parameters) and turbo distillation versions for faster inference.
  • ·Multi-view image-to-3D generation for improved geometric consistency.
  • ·Official Blender add-on to integrate 3D generation directly into the modelling software.
  • ·Locally hostable Gradio interface for use without writing code.
  • ·API server available for integrating generation into custom pipelines.
  • ·Community integration with ComfyUI via the ComfyUI-Hunyuan3DWrapper and ComfyUI-3D-Pack extensions.

In practice

Price
Free
Commercial use
Commercial use: no
Available on
linux, windows, macos
Host
comfyui
Licence
Tencent Hunyuan Community
Runs locally
Yes, 12 GB VRAM recommended
FR interface
No

Security and data

Rating
Security: one caveat, see the card
Your data
Data stays local
Maintenance
Inactive for over six months

Links and repository checked on 25 October 2026 · added on 1 January 2025

How to install Hunyuan3D 2.1 (Tencent)

1. Install PyTorch from the official site (pytorch.org), choosing the version suited to your system and GPU. 2. Clone the GitHub repository: git clone https://github.com/Tencent-Hunyuan/Hunyuan3D-2 3. Install Python dependencies: pip install -r requirements.txt && pip install -e . 4. Compile the rasterisation module: cd hy3dgen/texgen/custom_rasterizer && python3 setup.py install 5. Compile the differentiable renderer: cd hy3dgen/texgen/differentiable_renderer && python3 setup.py install 6. On Windows, you can use the community tool Hunyuan3D-2-WinPortable to avoid manual compilation steps.
pip install -r requirements.txt && pip install -e .

In the same category

Whole category 3D & printing →