You're looking at a specific version of this model. Jump to the model overview.
enhancerandroid02 /flix_upscaler_1.0.0:be468bf8
Input schema
The fields you can use to run this model with an API. If you don’t give a value for a field its default value will be used.
| Field | Type | Default value | Description |
|---|---|---|---|
| video |
string
|
Input video (.mp4/.mov/.webm)
|
|
| scale |
integer
|
2
Min: 2 Max: 4 |
Upscale factor
|
| keep_audio |
boolean
|
False
|
Preserve audio track if present
|
| tile_dit |
boolean
|
False
|
Enable tiled DiT inference (reduces VRAM usage)
|
| tile_size |
integer
|
256
Min: 16 Max: 2048 |
Tile size for tiling (DiT pixel tile size)
|
| overlap |
integer
|
24
Max: 1024 |
Tile overlap (reduces seams)
|
| fps |
integer
|
0
Max: 60 |
Force output FPS. Use 0 to match input (or 30 for image sequences).
|
| input_size_ratio |
number
|
1
Min: 0.1 Max: 1 |
Downscale input frames before upscaling for faster inference (0.1-1.0).
|
| quality |
integer
|
10
Max: 10 |
Output video quality (0-10)
|
| negative_prompt |
string
|
|
Negative prompt
|
| seed |
integer
|
0
|
Random seed
|
| sparse_ratio |
number
|
1.5
Min: 0.5 Max: 5 |
Sparse attention ratio (1.5=faster, 2.0=stable)
|
| kv_ratio |
number
|
3
Min: 0.5 Max: 10 |
KV cache ratio
|
| local_range |
integer
|
11
Min: 1 Max: 64 |
Local attention range (9=sharper, 11=stable)
|
| color_fix |
boolean
|
False
|
Apply color correction
|
Output schema
The shape of the response you’ll get when you run this model with an API.
{'format': 'uri', 'title': 'Output', 'type': 'string'}