---
license: other
license_name: flux-1-dev-non-commercial-license
license_link: LICENSE.md
datasets:
- Lakonik/t2i-prompts-3m
base_model:
- black-forest-labs/FLUX.1-dev
pipeline_tag: text-to-image
library_name: diffusers
---

# pi-Flow: Policy-Based Flow Models

Distilled 4-step and 8-step FLUX.1 models proposed in the paper:

**pi-Flow: Policy-Based Few-Step Generation via Imitation Distillation**
<br>
[Hansheng Chen](https://lakonik.github.io/)<sup>1</sup>, 
[Kai Zhang](https://kai-46.github.io/website/)<sup>2</sup>,
[Hao Tan](https://research.adobe.com/person/hao-tan/)<sup>2</sup>,
[Leonidas Guibas](https://geometry.stanford.edu/?member=guibas)<sup>1</sup>,
[Gordon Wetzstein](http://web.stanford.edu/~gordonwz/)<sup>1</sup>, 
[Sai Bi](https://sai-bi.github.io/)<sup>2</sup><br>
<sup>1</sup>Stanford University, <sup>2</sup>Adobe Research
<br>
[[arXiv](https://arxiv.org/abs/2510.14974)] [[Code](https://github.com/Lakonik/piFlow)] [[pi-Qwen Demo🤗](https://huggingface.co/spaces/Lakonik/pi-Qwen)] [[pi-FLUX Demo🤗](https://huggingface.co/spaces/Lakonik/pi-FLUX.1)] 

![teaser](https://cdn-uploads.huggingface.co/production/uploads/638067fcb334960c987fbeda/H0J1LYUcSS5YqOwZqQ0Jb.jpeg)

## Usage

Please first install the [official code repository](https://github.com/Lakonik/piFlow).

We provide diffusers pipelines for easy inference. The following code demonstrates how to sample images from the distilled FLUX models.

### 4-NFE GM-FLUX (GMFlow Policy)
Note: For the 8-NFE version, replace `gmflux_k8_piid_4step` with `gmflux_k8_piid_8step` and set `num_inference_steps=8`.
```python
import torch
from diffusers import FlowMatchEulerDiscreteScheduler
from lakonlab.pipelines.piflux_pipeline import PiFluxPipeline

pipe = PiFluxPipeline.from_pretrained(
    'black-forest-labs/FLUX.1-dev',
    torch_dtype=torch.bfloat16)
adapter_name = pipe.load_piflow_adapter(  # you may later call `pipe.set_adapters([adapter_name, ...])` to combine other adapters (e.g., style LoRAs)
    'Lakonik/pi-FLUX.1',
    subfolder='gmflux_k8_piid_4step',
    target_module_name='transformer')
pipe.scheduler = FlowMatchEulerDiscreteScheduler.from_config(  # use fixed shift=3.2
    pipe.scheduler.config, shift=3.2, use_dynamic_shifting=False)
pipe = pipe.to('cuda')

out = pipe(
    prompt='A portrait photo of a kangaroo wearing an orange hoodie and blue sunglasses standing in front of the Sydney Opera House holding a sign on the chest that says "Welcome Friends"',
    width=1360,
    height=768,
    num_inference_steps=4,
    generator=torch.Generator().manual_seed(42),
).images[0]
out.save('gmflux_4nfe.png')
```

![gmflux_4nfe](https://cdn-uploads.huggingface.co/production/uploads/638067fcb334960c987fbeda/KJNf9OJjGj5tbCq_WmAIR.png)

### 4-NFE DX-FLUX (DX Policy)
```
import torch
from diffusers import FlowMatchEulerDiscreteScheduler
from lakonlab.pipelines.piflux_pipeline import PiFluxPipeline

pipe = PiFluxPipeline.from_pretrained(
    'black-forest-labs/FLUX.1-dev',
    policy_type='DX',
    policy_kwargs=dict(
        segment_size=1 / 3.5,  # 1 / (nfe - 1 + final_step_size_scale)
        shift=3.2),
    torch_dtype=torch.bfloat16)
adapter_name = pipe.load_piflow_adapter(  # you may later call `pipe.set_adapters([adapter_name, ...])` to combine other adapters (e.g., style LoRAs)
    'Lakonik/pi-FLUX.1',
    subfolder='dxflux_n10_piid_4step',
    target_module_name='transformer')
pipe.scheduler = FlowMatchEulerDiscreteScheduler.from_config(  # use fixed shift=3.2
    pipe.scheduler.config, shift=3.2, use_dynamic_shifting=False)
pipe = pipe.to('cuda')

out = pipe(
    prompt='A portrait photo of a kangaroo wearing an orange hoodie and blue sunglasses standing in front of the Sydney Opera House holding a sign on the chest that says "Welcome Friends"',
    width=1360,
    height=768,
    num_inference_steps=4,
    generator=torch.Generator().manual_seed(42),
).images[0]
out.save('dxflux_4nfe.png')
```

![dxflux_4nfe](https://cdn-uploads.huggingface.co/production/uploads/638067fcb334960c987fbeda/sSNJFtEecY41TsuciQQFL.png)

## Citation
```
@misc{piflow,
      title={pi-Flow: Policy-Based Few-Step Generation via Imitation Distillation}, 
      author={Hansheng Chen and Kai Zhang and Hao Tan and Leonidas Guibas and Gordon Wetzstein and Sai Bi},
      year={2025},
      eprint={2510.14974},
      archivePrefix={arXiv},
      primaryClass={cs.LG},
      url={https://arxiv.org/abs/2510.14974}, 
}
```