Try MiniMax H3 (Text-to-Video) (Image-to-Video) from MiniMax →
Sign InStart Building

Task History

Click to see output list

No tasks yet

Go to Models
Image GenerationActive

wiro / train-flux-lora

train-flux-lora

bywiro

LoRA training is a low-rank adaptation technique used to fine-tune FLUX models, enabling personalized model outputs with minimal computational overhead.

Text to ImageTrainFlux
Model ID
train-flux-lora
Provider
wiro
Added
1734698741
wiro playground—wiro/train-flux-lora
Step 1
Train Details
Step 2
Train Advance Details
Step 3
Upload / Choose Train Data
Reset to defaults

Your new model will be named this.

You can write your model description to identify it in the future.

Your model will be trained on this reference model.

If you check the Quantize checkbox, the base model will be loaded in a quantized format, which significantly reduces VRAM usage and allows training on low-VRAM GPUs. However, this may result in a slight decrease in output quality.

Added 1734698741

API quick start

Run train-flux-lora with a single API call.

POST https://api.wiro.ai/v1/Run/wiro/train-flux-lora
{
  "prompt": "ohwx, portrait of ohwx",
  "trainModelName": "...",
  "referenceModel": "...",
  "triggerWord": "ohwx"
}
View full API docs