---
library_name: diffusers
tags:
- text-to-image
- sana
- one-step
- gpu
- fastapi
---

# sana_bullet

Fast finetuned one-step Sana checkpoint for the standalone `sana_bullet` GPU inference stack.

## Links

- GitHub repo: [rotem154154/sana_bullet](https://github.com/rotem154154/sana_bullet)
- Model repo: [irotem98/sana_bullet](https://huggingface.co/irotem98/sana_bullet)

## One-Line Generation

```python
# pip install "git+https://github.com/rotem154154/sana_bullet.git"
from sana_bullet import generate_image

image = generate_image(
    "a cinematic portrait of a fox in soft winter light",
    model="irotem98/sana_bullet",
)
image
```

## `from_pretrained(...)` Loading

```python
# pip install "git+https://github.com/rotem154154/sana_bullet.git"
from sana_bullet import SanaBulletPipeline

pipe = SanaBulletPipeline.from_pretrained("irotem98/sana_bullet")
image = pipe.generate("a retro-futuristic motorcycle speeding through neon rain")
image
```

## Server Usage

```bash
pip install "git+https://github.com/rotem154154/sana_bullet.git"
export SANA_BULLET_CHECKPOINT_REPO_ID="irotem98/sana_bullet"
python -m sana_bullet.ui_server
```

## Files

- `step0003000.pt`: finetuned transformer checkpoint

## Base Stack

- Base model: `Efficient-Large-Model/Sana_Sprint_1.6B_1024px_diffusers`
- VAE: `dc-ai/dc-ae-lite-f32c32-diffusers`

## Notes

- The checkpoint is designed for GPU inference.
- The higher-level Python API lives in the GitHub repo, while this model repo only hosts the weights and model card.