
Pixel Art Style Lora Z Image Turbo GGUF
Note: All model files and custom nodes below must be downloaded via the client. If you do not have the client, Download >>>
Pixel Art Style Lora for Z Image Turbo (GGUF format). You should use "Pixel art style" to trigger the image generation.
| Name | Size | Target Dir |
|---|
| ae.safetensors | 335 MB | vae |
| Name | Size | Target Dir |
|---|
| pixel_art_style_z_image_turbo.safetensors | 170.13 MB | loras |
| Name | Size | Target Dir |
|---|
| z_image_turbo-Q3_K_S.gguf | 3.79 GB | unet |
| z_image_turbo-Q3_K_M.gguf | 4.12 GB | unet |
| z_image_turbo-Q4_K_S.gguf | 4.66 GB | unet |
| z_image_turbo-Q4_K_M.gguf | 4.98 GB | unet |
| z_image_turbo-Q5_K_S.gguf | 5.19 GB | unet |
| z_image_turbo-Q5_K_M.gguf | 5.52 GB | unet |
| z_image_turbo-Q6_K.gguf | 5.91 GB | unet |
| z_image_turbo-Q8_0.gguf | 7.22 GB | unet |
| Name | Size | Target Dir |
|---|
| Qwen3-4B-Q4_K_M.gguf | 2.50 GB | text_encoders |
| Qwen3-4B-Q5_0.gguf | 2.82 GB | text_encoders |
| Qwen3-4B-Q5_K_M.gguf | 2.89 GB | text_encoders |
| Qwen3-4B-Q6_K.gguf | 3.31 GB | text_encoders |
| Qwen3-4B-Q8_0.gguf | 4.28 GB | text_encoders |