Model card auto-generated by SimpleTuner

README.md

@@ -1,6 +1,6 @@
 ---
 license: other
-base_model: "black-forest-labs/FLUX.1-schnell"
+base_model: "black-forest-labs/FLUX.1-dev"
 tags:
   - flux
   - flux-diffusers
@@ -25,7 +25,7 @@ widget:
 
 # lora-training
 
-This is a standard PEFT LoRA derived from [black-forest-labs/FLUX.1-schnell](https://huggingface.co/black-forest-labs/FLUX.1-schnell).
+This is a standard PEFT LoRA derived from [black-forest-labs/FLUX.1-dev](https://huggingface.co/black-forest-labs/FLUX.1-dev).
 
 
 The main validation prompt used during training was:
@@ -58,7 +58,7 @@ You may reuse the base model text encoder for inference.
 ## Training settings
 
 - Training epochs: 0
-- Training steps: 300
+- Training steps: 100
 - Learning rate: 0.0008
 - Effective batch size: 1
   - Micro-batch size: 1
@@ -70,7 +70,7 @@ You may reuse the base model text encoder for inference.
 - Precision: bf16
 - Quantised: Yes: int8-quanto
 - Xformers: Not used
-- LoRA Rank: 16
+- LoRA Rank: 8
 - LoRA Alpha: None
 - LoRA Dropout: 0.1
 - LoRA initialisation style: default
@@ -80,7 +80,7 @@ You may reuse the base model text encoder for inference.
 
 ### right-triangles
 - Repeats: 0
-- Total number of images: 348
+- Total number of images: 380
 - Total number of aspect buckets: 1
 - Resolution: 512 px
 - Cropped: True
@@ -95,7 +95,7 @@ You may reuse the base model text encoder for inference.
 import torch
 from diffusers import DiffusionPipeline
 
-model_id = 'black-forest-labs/FLUX.1-schnell'
+model_id = 'black-forest-labs/FLUX.1-dev'
 adapter_id = 'Mujeeb603/lora-training'
 pipeline = DiffusionPipeline.from_pretrained(model_id)
 pipeline.load_lora_weights(adapter_id)