---
license: other
base_model: "black-forest-labs/FLUX.2-dev"
tags:
  - flux2
  - flux2-diffusers
  - text-to-image
  - image-to-image
  - diffusers
  - simpletuner
  - not-for-all-audiences
  - lora

  - template:sd-lora
  - standard
pipeline_tag: text-to-image
inference: true

---

# quzo/fl2

This is a PEFT LoRA derived from [black-forest-labs/FLUX.2-dev](https://huggingface.co/black-forest-labs/FLUX.2-dev).

The main validation prompt used during training was:
```
bm82 man
```


## Validation settings
- CFG: `7.5`
- CFG Rescale: `0.0`
- Steps: `20`
- Sampler: `FlowMatchEulerDiscreteScheduler`
- Seed: `None`
- Resolution: `1024x1024`


Note: The validation settings are not necessarily the same as the [training settings](#training-settings).




<Gallery />

The text encoder **was not** trained.
You may reuse the base model text encoder for inference.


## Training settings

- Training epochs: 566
- Training steps: 3400
- Learning rate: 0.0001
  - Learning rate schedule: constant_with_warmup
  - Warmup steps: 0
- Max grad value: 2.0
- Effective batch size: 2
  - Micro-batch size: 2
  - Gradient accumulation steps: 1
  - Number of GPUs: 1
- Gradient checkpointing: True
- Prediction type: flow_matching[]
- Optimizer: adamw_bf16
- Trainable parameter precision: Pure BF16
- Base model precision: `no_change`
- Caption dropout probability: 0.1%



- LoRA Rank: 16
- LoRA Alpha: 16.0
- LoRA Dropout: 0.1
- LoRA initialisation style: default
- LoRA mode: Standard
    

## Datasets

### training-images
- Repeats: 0
- Total number of images: 12
- Total number of aspect buckets: 1
- Resolution: 1.048576 megapixels
- Cropped: False
- Crop style: None
- Crop aspect: None
- Used for regularisation data: No


## Inference


```python
import torch
from diffusers import DiffusionPipeline

model_id = 'black-forest-labs/FLUX.2-dev'
adapter_id = 'quzo/fl2'
pipeline = DiffusionPipeline.from_pretrained(model_id, torch_dtype=torch.bfloat16) # loading directly in bf16
pipeline.load_lora_weights(adapter_id)

prompt = "bm82 man"
negative_prompt = 'blurry, cropped, ugly'

## Optional: quantise the model to save on vram.
## Note: The model was not quantised during training, so it is not necessary to quantise it during inference time.
#from optimum.quanto import quantize, freeze, qint8
#quantize(pipeline.transformer, weights=qint8)
#freeze(pipeline.transformer)
    
pipeline.to('cuda' if torch.cuda.is_available() else 'mps' if torch.backends.mps.is_available() else 'cpu') # the pipeline is already in its target precision level
model_output = pipeline(
    prompt=prompt,
    negative_prompt=negative_prompt,
    num_inference_steps=20,
    generator=torch.Generator(device='cuda' if torch.cuda.is_available() else 'mps' if torch.backends.mps.is_available() else 'cpu').manual_seed(42),
    width=1024,
    height=1024,
    guidance_scale=7.5,
).images[0]

model_output.save("output.png", format="PNG")

```