Configuring a LoRA training job on RunComfy
RunComfyDemo summary
The demonstration covers setting up a training job for the Z-Image architecture, including setting the trigger word, step count, and sample prompts for monitoring progress.
Step-by-step
- Go to New Job in the training menu
- Enter a name for the job and set the Trigger Word
- Select the model architecture (e.g., Z-Image)
- Adjust the total steps (e.g., 4,000 steps)
- Select the uploaded data set
- Enter sample prompts using the trigger word to monitor training progress
- Click Create Job
Options
- Toggle Low VRAM mode
- Adjust the interval for saving tensor files (default is every 250 steps)
- Modify various advanced settings for specific job requirements
Watch out for
- You must select the correct model architecture to start the training
Tips
- Leave most settings at their defaults unless you have a specific reason to change them
- Use the same sample prompt across multiple slots to consistently gauge how well the LoRA is training over time
- Ensure the trigger word is included in your sample prompts to test the LoRA's activation
All demos from “Train a LoRA for LTX 2.3 / Z-Image on RunComfy | Dataset Captioning + AI Toolkit Tutorial”
- 6:381:58Creating a new dataset in RunComfy TrainerThe user shows how to upload a zipped folder of images and matching text captions to the RunComfy Trainer interface to define a new training dataset.RunComfy· AI Avatar Generator
- 8:362:41Configuring a LoRA training job on RunComfyCurrentThe demonstration covers setting up a training job for the Z-Image architecture, including setting the trigger word, step count, and sample prompts for monitoring progress.RunComfy· AI Avatar Generator
- 13:251:10Reviewing LoRA training sample resultsThe creator reviews the output samples from the training process, comparing different step intervals to identify the best art style replication for the D&D monster LoRA.RunComfy· AI Avatar Generator
- Watch “Train a LoRA for LTX 2.3 / Z-Image on RunComfy | Dataset Captioning + AI Toolkit Tutorial” →
AI Avatar Generator
- 6:381:58Creating a new dataset in RunComfy TrainerThe user shows how to upload a zipped folder of images and matching text captions to the RunComfy Trainer interface to define a new training dataset.Eye For AI
- 8:362:41Configuring a LoRA training job on RunComfyCurrentThe demonstration covers setting up a training job for the Z-Image architecture, including setting the trigger word, step count, and sample prompts for monitoring progress.Eye For AI
- 13:251:10Reviewing LoRA training sample resultsThe creator reviews the output samples from the training process, comparing different step intervals to identify the best art style replication for the D&D monster LoRA.Eye For AI