Licenses & Third-Party Attributions
DiffusionBear is released under the MIT License. Model weights are never bundled — they are downloaded directly from Hugging Face or Civitai under their respective upstream licenses. This page is synchronized with the in-app Parameters → Licences tab.
Model Family Attributions
DiffusionBear uses Qwen language and image models for on-device prompt enhancement and generation. Qwen is developed by Hangzhou Tongyi Laboratory Technology Co., Ltd.
Qwen-Image 2.1 and Turbo are licensed under the Qwen RESEARCH LICENSE AGREEMENT, Copyright (c) 2026 Hangzhou Tongyi Laboratory Technology Co., Ltd. All Rights Reserved. Qwen-Image Edit and Qwen2.5 Instruct are Apache-2.0.
Image generation capabilities are powered by models from the OpenMOSS Team (Apache License 2.0).
Advanced generation and editing powered by Bria AI models (Creative Commons CC BY-NC 4.0).
Supported Models & Licenses
| Model | Upstream License | Commercial Outputs | Commercial Weights |
|---|---|---|---|
| FLUX.2-klein 4B | Apache-2.0 | Yes | Yes |
| FLUX.2-klein 9B | FLUX Non-Commercial License | Yes | No |
| FLUX.3-klein 5.6B | FLUX Kommunity License v1.0 | Conditional | Conditional |
| FLUX.1-schnell | Apache-2.0 | Yes | Yes |
| MOSS-Image Base 5B | Apache-2.0 | Yes | Yes |
| MOSS-Image SFT 5B | Apache-2.0 | Yes | Yes |
| Kromoss 5B | Apache-2.0 | Yes | Yes |
| Z-Image Turbo 6B | Apache-2.0 | Yes | Yes |
| Qwen-Image Edit | Apache-2.0 | Yes | Yes |
| Qwen-Image 2.1 (MLX q4) | Qwen Research License | No | No |
| Qwen-Image 2.1 Turbo (MLX q4) | Qwen Research License | No | No |
| Krea 2 Turbo 13B | Krea 2 Community License | Conditional | Conditional |
| FIBO 2.0 Edit Turbo | CC BY-NC 4.0 | No | No |
| Bria RMBG-3.0 (Background Removal) | CC BY-NC 4.0 | No | No |
| Juggernaut XL Lightning (MLX) | CreativeML OpenRAIL-M | Yes | Conditional |
| RealVisXL V5.0 Lightning (MLX) | OpenRAIL++ | Yes | Yes |
| RealVisXL V5.0 Standard / Hyper-SD | OpenRAIL++ | Yes | Yes |
| Juggernaut XI v11 (MLX) | CC BY-NC-ND 4.0 | No | No |
| Pony Diffusion V6 XL | Modified Fair AI Public License 1.0-SD | Yes | Conditional |
| Qwen2.5-1.5B-Instruct (prompt enhancer) | Apache-2.0 | Yes | Yes |
| Qwen2.5-0.5B-Instruct (prompt enhancer) | Apache-2.0 | Yes | Yes |