thefluxtrain/omini-kontext
An inference and training framework for multiple image input in Flux Kontext dev
Run thefluxtrain/omini-kontext with an API
Use one of our client libraries to get started quickly. Clicking on a library will take you to the Playground tab where you can tweak different inputs, see the results, and copy the corresponding code to use in your own project.
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 |
|---|---|---|---|
| image |
string
|
Image to insert character into
|
|
| reference_image |
string
|
Reference image
|
|
| task |
None
|
character_insertion
|
Task
|
| delta |
string
|
[1, 0, 0]
|
Reference delta
|
| prompt |
string
|
Add character to the scene
|
Input prompt.
|
| negative_prompt |
string
|
Bad, low quality, deformed, distorted, distorted, ugly
|
Input negative prompt.
|
| should_optimise_reference |
boolean
|
Should optimise reference?
|
|
| num_inference_steps |
integer
|
20
Min: 1 Max: 150 |
Number of denoising steps
|
| lora_strength |
number
|
0.6
Max: 1 |
LoRA strength
|
| guidance_scale |
number
|
4.5
Max: 10 |
Text guidance scale
|
| seed |
integer
|
Random seed. Leave blank to randomize the seed
|
|
| lora_path |
string
|
HF path to the LoRA weights, if custom task is choosen
|
|
| lora_weight_name |
string
|
Weight name of the LoRA weights, if custom task is choosen
|
{
"type": "object",
"title": "Input",
"properties": {
"seed": {
"type": "integer",
"title": "Seed",
"x-order": 10,
"description": "Random seed. Leave blank to randomize the seed"
},
"task": {
"enum": [
"character_insertion",
"spatial_character_insertion",
"product_insertion",
"none",
"custom"
],
"type": "string",
"title": "task",
"description": "Task",
"default": "character_insertion",
"x-order": 2
},
"delta": {
"type": "string",
"title": "Delta",
"default": "[1, 0, 0]",
"x-order": 3,
"description": "Reference delta"
},
"image": {
"type": "string",
"title": "Image",
"format": "uri",
"x-order": 0,
"description": "Image to insert character into"
},
"prompt": {
"type": "string",
"title": "Prompt",
"default": "Add character to the scene",
"x-order": 4,
"description": "Input prompt."
},
"lora_path": {
"type": "string",
"title": "Lora Path",
"x-order": 11,
"description": "HF path to the LoRA weights, if custom task is choosen"
},
"lora_strength": {
"type": "number",
"title": "Lora Strength",
"default": 0.6,
"maximum": 1,
"minimum": 0,
"x-order": 8,
"description": "LoRA strength"
},
"guidance_scale": {
"type": "number",
"title": "Guidance Scale",
"default": 4.5,
"maximum": 10,
"minimum": 0,
"x-order": 9,
"description": "Text guidance scale"
},
"negative_prompt": {
"type": "string",
"title": "Negative Prompt",
"default": "Bad, low quality, deformed, distorted, distorted, ugly",
"x-order": 5,
"description": "Input negative prompt."
},
"reference_image": {
"type": "string",
"title": "Reference Image",
"format": "uri",
"x-order": 1,
"description": "Reference image"
},
"lora_weight_name": {
"type": "string",
"title": "Lora Weight Name",
"x-order": 12,
"description": "Weight name of the LoRA weights, if custom task is choosen"
},
"num_inference_steps": {
"type": "integer",
"title": "Num Inference Steps",
"default": 20,
"maximum": 150,
"minimum": 1,
"x-order": 7,
"description": "Number of denoising steps"
},
"should_optimise_reference": {
"type": "boolean",
"title": "Should Optimise Reference",
"x-order": 6,
"description": "Should optimise reference?"
}
}
}
Output schema
The shape of the response you’ll get when you run this model with an API.
{
"type": "string",
"title": "Output",
"format": "uri"
}