Type: mlx.image_to_image.MFluxDepth
Namespace: mlx.image_to_image
Description
Generate images with depth guidance via the MFlux depth pipeline using local MLX acceleration. mlx, flux, depth, conditioning, structure-preserving
Use cases:
- Use a depth map to control structural composition while keeping prompt-driven appearance
- Provide both source image and depth map to transfer scene layout to a new generation
- Generate depth-guided outputs when only a depth map is available (source image optional)
Properties
| Property | Type | Description | Default |
|---|---|---|---|
| prompt | str |
Primary text prompt for the depth-guided generation. | Highly detailed cinematic portrait with depth cues |
| image | image |
Optional reference image used for depth extraction or as a content guide. | {'type': 'image', 'uri': '', 'asset_id': None, 'data': None} |
| depth_image | image |
Optional depth map to guide geometry. If omitted, depth is inferred from the image when provided. | {'type': 'image', 'uri': '', 'asset_id': None, 'data': None} |
| model | hf.flux_depth |
Depth model weights compatible with the Flux depth pipeline. | {'type': 'hf.flux_depth', 'repo_id': 'black-forest-labs/FLUX.1-Depth-dev', 'path': None, 'variant': None, 'allow_patterns': None, 'ignore_patterns': None} |
| quantize | Optional[Enum[3, 4, 5, 6, 8]] |
Optional quantization level for model weights (reduces memory usage). | 4 |
| steps | int |
Number of denoising steps for the generation run. | 20 |
| guidance | Optional[float] |
Classifier-free guidance scale. Defaults higher to encourage prompt adherence in depth mode. | 10.0 |
| height | int |
Height of the generated image in pixels. | 1024 |
| width | int |
Width of the generated image in pixels. | 1024 |
| seed | int |
Seed for deterministic generation. Leave 0 for random seed. | 0 |
Outputs
| Output | Type | Description |
|---|---|---|
| output | image |
Metadata
Related Nodes
Browse other nodes in the mlx.image_to_image namespace.