Qwen-Image 8 Steps GGUF
Note: All model files and custom nodes below must be downloaded via the client. If you do not have the client, Download >>>
Qwen-Image is the first image generation foundation model in the Tongyi Qianwen series, released in August 2025. It has 20 billion parameters and performs outstandingly in text rendering, image generation, and editing. It excels at high-fidelity text rendering for multi-line paragraphs in both Chinese and English, and has strong processing capabilities for complex scenes and fine-grained details, accurately presenting text content. In multiple public benchmark tests, this model has achieved state-of-the-art performance, especially leading existing models by a large margin in Chinese text rendering. Additionally, it supports image generation in various styles and professional-level image editing operations, providing powerful assistance for image creation.
| Name | Size | Target Dir |
|---|---|---|
| qwen-image-Q2_K.gguf | 7.06 GB | unet |
| qwen-image-Q3_K_S.gguf | 8.95 GB | unet |
| qwen-image-Q3_K_M.gguf | 9.68 GB | unet |
| qwen-image-Q4_0.gguf | 11.85 GB | unet |
| qwen-image-Q4_K_S.gguf | 12.14 GB | unet |
| qwen-image-Q4_1.gguf | 12.84 GB | unet |
| qwen-image-Q4_K_M.gguf | 13.07 GB | unet |
| qwen-image-Q5_K_S.gguf | 14.12 GB | unet |
| qwen-image-Q5_0.gguf | 14.40 GB | unet |
| qwen-image-Q5_K_M.gguf | 14.93 GB | unet |
| qwen-image-Q5_1.gguf | 15.39 GB | unet |
| qwen-image-Q6_K.gguf | 16.82 GB | unet |
| qwen-image-Q8_0.gguf | 21.76 GB | unet |
| qwen-image-BF16.gguf | 40.87 GB | unet |
| Name | Size | Target Dir |
|---|---|---|
| Qwen-Image-Lightning-4steps-V2.0.safetensors | 1.70 GB | loras |
| Qwen-Image-Lightning-4steps-V2.0-bf16.safetensors | 850 MB | loras |
| Qwen-Image-Lightning-4steps-V1.0.safetensors | 1.70 GB | loras |
| Qwen-Image-Lightning-8steps-V1.0.safetensors | 1.70 GB | loras |
| Qwen-Image-Lightning-8steps-V1.1.safetensors | 1.70 GB | loras |
| Qwen-Image-Lightning-4steps-V1.0-bf16.safetensors | 849.61 MB | loras |
| Qwen-Image-Lightning-8steps-V1.1-bf16.safetensors | 849.61 MB | loras |
| Name | Size | Target Dir |
|---|---|---|
| qwen_image_vae.safetensors | 253.81 MB | vae |
| Name | Size | Target Dir |
|---|---|---|
| Qwen2.5-VL-7B-Instruct-UD-IQ1_S.gguf | 2.07 GB | text_encoders |
| Qwen2.5-VL-7B-Instruct-UD-IQ1_M.gguf | 2.20 GB | text_encoders |
| Qwen2.5-VL-7B-Instruct-UD-IQ2_XXS.gguf | 2.40 GB | text_encoders |
| Qwen2.5-VL-7B-Instruct-UD-IQ2_M.gguf | 2.86 GB | text_encoders |
| Qwen2.5-VL-7B-Instruct-Q2_K.gguf | 3.02 GB | text_encoders |
| Qwen2.5-VL-7B-Instruct-Q2_K_L.gguf | 3.14 GB | text_encoders |
| Qwen2.5-VL-7B-Instruct-UD-IQ3_XXS.gguf | 3.17 GB | text_encoders |
| Qwen2.5-VL-7B-Instruct-UD-Q2_K_XL.gguf | 3.24 GB | text_encoders |
| Qwen2.5-VL-7B-Instruct-Q3_K_S.gguf | 3.49 GB | text_encoders |
| Qwen2.5-VL-7B-Instruct-Q3_K_M.gguf | 3.81 GB | text_encoders |
| Qwen2.5-VL-7B-Instruct-UD-Q3_K_XL.gguf | 4.00 GB | text_encoders |
| Qwen2.5-VL-7B-Instruct-IQ4_XS.gguf | 4.24 GB | text_encoders |
| Qwen2.5-VL-7B-Instruct-IQ4_NL.gguf | 4.44 GB | text_encoders |
| Qwen2.5-VL-7B-Instruct-Q4_0.gguf | 4.44 GB | text_encoders |
| Qwen2.5-VL-7B-Instruct-Q4_K_S.gguf | 4.46 GB | text_encoders |
| Qwen2.5-VL-7B-Instruct-Q4_K_M.gguf | 4.68 GB | text_encoders |
| Qwen2.5-VL-7B-Instruct-UD-Q4_K_XL.gguf | 4.79 GB | text_encoders |
| Qwen2.5-VL-7B-Instruct-Q4_1.gguf | 4.87 GB | text_encoders |
| Qwen2.5-VL-7B-Instruct-Q5_K_S.gguf | 5.32 GB | text_encoders |
| Qwen2.5-VL-7B-Instruct-Q5_K_M.gguf | 5.44 GB | text_encoders |
| Qwen2.5-VL-7B-Instruct-UD-Q5_K_XL.gguf | 5.46 GB | text_encoders |
| Qwen2.5-VL-7B-Instruct-Q6_K.gguf | 6.25 GB | text_encoders |
| Qwen2.5-VL-7B-Instruct-UD-Q6_K_XL.gguf | 6.96 GB | text_encoders |
| Qwen2.5-VL-7B-Instruct-Q8_0.gguf | 8.10 GB | text_encoders |
| Qwen2.5-VL-7B-Instruct-UD-Q8_K_XL.gguf | 10.19 GB | text_encoders |
| Qwen2.5-VL-7B-Instruct-BF16.gguf | 15.24 GB | text_encoders |