Flora (FGO)

⚠️ Important Note: Experimental LoRA (Test Version)

This model is currently in the testing phase. Please follow these guidelines for the best experience:

• Instability: The model may "hallucinate" or produce artifacts with overly complex prompts.

• Recommended Weights: We suggest using a strength between 0.6 and 0.8 to prevent distortions.

• Sampling Method: Best results are achieved using DPM++ 2M.

• Compatibility: This LoRA supports all checkpoints and can be mixed with other LoRAs without issues.

Work in Progress: We are still refining the dataset. If the output looks strange, it is because the model is still learning specific patterns. Use at your own risk, push the limits, and please report any issues in the comments! 🚀