Image: Enhance

qwen-image-lora-trainer (balanced)

Train custom Qwen-Image LoRA models from a dataset zip URL. Input: dataset zip URL, optional trigger_word, steps, learning_rate, lora_rank. Output: trained LoRA model weights URL. Use for: qwen lora training, fine-tuning, custom style training, character lora.

post/api/v1/qwen-image-lora-trainer

Request

  • Base URL: https://api.muapi.ai
  • URL: https://api.muapi.ai/api/v1/qwen-image-lora-trainer
  • Auth: one of:
    • API key in header x-api-key
    • OAuth 2 (client credentials, token URL https://api.muapi.ai/oauth/token)

Query parameters

webhookstring
team_idinteger nullable

Headers

x-api-keystring
x-embed-codestring nullable
x-agent-embed-codestring nullable
x-whitelabel-studiostring nullable
authorizationstring nullable

Request body

datastring uri required

URL to zip archive with images. Try to use at least 4 images in general the more the better. In addition to images the archive can contain text files with captions.

trigger_wordstring nullable

Optional trigger word. If a caption file exists, it is prepended when not already present. Leave empty to enable no-caption training.

stepsinteger nullable

Number of steps to train the LoRA on.

learning_ratenumber nullable

Learning rate for Qwen LoRA training.

lora_rankinteger nullable

LoRA rank dimension.

webhook_urlstring uri nullable

Response

Successful Response

unknown required

Changes

No recorded changes to this operation across all 2 revisions of this API.