awesome-genmedia Pruna | P-Image LoRA | Text to Image
Generate images from text descriptions using custom LoRA weights from HuggingFace. Supports configurable aspect ratios, prompt upsampling, and custom dimensions.
Quick Start
Requires an each::labs API key. Get one at eachlabs.ai.
curl -X POST https://api.eachlabs.ai/v1/prediction \
-H "Content-Type: application/json" \
-H "X-API-Key: $EACHLABS_API_KEY" \
-d '{
"model": "p-image-lora-text-to-image",
"version": "0.0.1",
"input": {
"prompt": "A serene mountain landscape at sunset with golden light",
"lora_weights": "https://huggingface.co/my-org/my-style-lora",
"lora_scale": 0.5,
"aspect_ratio": "16:9"
}
}'
Parameters
| Parameter | Type | Default | Description |
|---|---|---|---|
| aspect_ratio | string | 16:9 | Aspect ratio for the generated image. custom: use width/height for exact dimensions. enum: 1:1, 16:9, 9:16, 4:3, 3:4, 3:2, 2:3, custom |
| disable_safety_checker | boolean | false | Disable safety checker for generated images. |
| height | integer | false | Custom height in pixels (256-1440, multiple of 16). Only when aspect_ratio=custom. |
| hf_api_token | string | false | HuggingFace API token for accessing private LoRA repositories. |
| lora_scale | number | 0.5 | LoRA strength (-1 to 3). 0.5 works well for most LoRAs. |
| lora_weights | string | false | HuggingFace URL to LoRA weights. |
| prompt | string | false | Text description of the image to generate. |
| prompt_upsampling | boolean | false | Upsample prompt with LLM for enhanced results. |
| seed | integer | false | Random seed for reproducible generation. |
| width | integer | false | Custom width in pixels (256-1440, multiple of 16). Only when aspect_ratio=custom. |
Examples
Generate a landscape image:
curl -X POST https://api.eachlabs.ai/v1/prediction \
-H "Content-Type: application/json" \
-H "X-API-Key: $EACHLABS_API_KEY" \
-d '{
"model": "p-image-lora-text-to-image",
"version": "0.0.1",
"input": {
"prompt": "A cyberpunk cityscape with neon lights reflecting on wet streets",
"lora_weights": "https://huggingface.co/my-org/cyberpunk-lora",
"aspect_ratio": "16:9",
"prompt_upsampling": true
}
}'
Custom dimensions with seed:
curl -X POST https://api.eachlabs.ai/v1/prediction \
-H "Content-Type: application/json" \
-H "X-API-Key: $EACHLABS_API_KEY" \
-d '{
"model": "p-image-lora-text-to-image",
"version": "0.0.1",
"input": {
"prompt": "Portrait of a fantasy elf warrior in armor",
"lora_weights": "https://huggingface.co/my-org/fantasy-lora",
"lora_scale": 0.7,
"aspect_ratio": "custom",
"width": 768,
"height": 1024,
"seed": 12345
}
}'
Square image with high LoRA influence:
curl -X POST https://api.eachlabs.ai/v1/prediction \
-H "Content-Type: application/json" \
-H "X-API-Key: $EACHLABS_API_KEY" \
-d '{
"model": "p-image-lora-text-to-image",
"version": "0.0.1",
"input": {
"prompt": "A still life painting of flowers in a vase",
"lora_weights": "https://huggingface.co/my-org/oil-painting-lora",
"lora_scale": 1.5,
"aspect_ratio": "1:1"
}
}'
Related Models
- p-image-edit-lora-image-edit - Image editing with LoRA weights
- p-image-text-to-image - Standard text to image without LoRA
- p-image-edit - Image editing without LoRA