---
license: other
license_name: flux-non-commercial-license
license_link: LICENSE.md
language:
  - en
datasets:
  - openvivo/VINS-120K
base_model: black-forest-labs/FLUX.2-klein-9B
library_name: diffusers
pipeline_tag: image-to-image
metrics:
  - vins-4keval-plus-plus-overall
  - vins-4keval-plus-plus-nc
  - imgedit-bench-score
  - cie-bench-total
tags:
  - image-editing
  - instruction-based-image-editing
  - 4k
  - distillation
  - dcae
---

# BluePixel-Turbo

**Native 4K instruction-based image editing in 4 steps.**

[Project Page](https://tech.vivo.com/BluePixel/) · [Paper](https://arxiv.org/abs/2610.19882) · [Code](https://github.com/vivo/BluePixel) · [Dataset](https://huggingface.co/datasets/openvivo/VINS-120K)

BluePixel-Turbo is the distilled image editing model from **BluePixel: Toward Native 4K Agentic Image Editing**, developed by vivo. It builds on [FLUX.2-klein-9B](https://huggingface.co/black-forest-labs/FLUX.2-klein-9B) with a DCAE-F64 autoencoder for efficient native 4K editing.

- **Native 4K output** without cascaded super-resolution.
- **4-step inference** through source-anchored target projection (SATP), distilled from a 50-step teacher.
- **Long-tail editing skills** learned through ACE-MOPD, followed by reinforcement learning for aesthetics and instruction fidelity.
- **8.7 seconds per 4K edit** on a single Kunlunxin P800 XPU (BF16), as reported on the project page. Latency varies by hardware.

## Updates

**2026.11.03** — We update BluePixel to the **SATP-distilled version**, leveraging teacher–student distillation to enable high-quality image editing with significantly fewer sampling steps and lower inference costs. This update is an intermediate checkpoint without additional post-training. Further optimization, including reinforcement learning, is planned to improve aesthetic quality and instruction-following fidelity.

## Quick Start

Use the custom Diffusers-based pipeline provided in the code repository:

```bash
git clone https://github.com/vivo/BluePixel.git
cd BluePixel
pip install -r requirements.txt
pip install --upgrade diffusers huggingface_hub
```

Run from the cloned repository in a CUDA environment with BF16 support:

```python
import torch
from PIL import Image
from huggingface_hub import snapshot_download
from vins_infer import load_pipeline

model_dir = snapshot_download("openvivo/BluePixel-Turbo")
pipe = load_pipeline(
    model_dir=model_dir,
    device="cuda:0",
    vae_tiling=False,  # Enable to reduce autoencoder memory usage.
    lora_scale=None,
    shift=0.0,
    mu_mode="empirical",
)

image = Image.open("assets/test_input.jpg").convert("RGB")
result = pipe(
    image=image,
    instruction="Make the sky sunset orange and add warm light.",
    num_inference_steps=4,
    guidance_scale=3.0,
    target_pixels=4096 * 4096,
    generator=torch.Generator(device="cuda:0").manual_seed(42),
).images[0]
result.save("output.png")
```

The pipeline approximately preserves the input aspect ratio and aligns dimensions to multiples of 64; non-square inputs produce non-square outputs. Use a Diffusers version that includes the FLUX.2 components imported by `vins_infer.py`.

## Model Versions

This repository tracks the latest released checkpoint of BluePixel-Turbo. Earlier revisions are preserved in the commit history of this repository and their model card snapshots remain reachable through the Hugging Face revision browser.

| Revision | Transformer weights | Notes |
| :--- | :--- | :--- |
| v1 (previous) | `ckpt800` SFT checkpoint | Initial public release of this repository |
| v2 (current) | `ckpt35` (STEP-35) checkpoint | Updated transformer weights; text encoder, VAE and tokenizer unchanged |

### What changed in v2

Only the **transformer** weights were retrained and updated:

- `transformer/diffusion_pytorch_model-*.safetensors` — all 10 shards replaced (new SHA-256 OIDs).
- `transformer/config.json`, `scheduler/scheduler_config.json`, `vae/config.json`, `transformer/diffusion_pytorch_model.safetensors.index.json` — metadata refresh (diffusers version bump, `torch_dtype` field, `_name_or_path` provenance).

The following components are **byte-identical** to the previous release and were not re-uploaded:

- `text_encoder/model-*.safetensors` (5 shards) and `text_encoder/*.json`
- `vae/diffusion_pytorch_model.safetensors`
- `tokenizer/tokenizer.json`, `tokenizer/tokenizer_config.json`, `tokenizer/chat_template.jinja`

The model architecture is unchanged: the transformer still holds 233 tensors with an identical `total_size` of 36,314,324,992 bytes, so this is a pure fine-tune weight update with no config or shape migration required.

> **Note on `torch_dtype`.** The transformer `config.json` now records `"torch_dtype": "bfloat16"`. The published weights themselves are stored in float32, matching the source checkpoint; the field describes the intended inference dtype.

## Benchmarks

Results below are reported for **BluePixel** on the [project page](https://tech.vivo.com/BluePixel/); the page does not assign each quality score specifically to this Turbo checkpoint.

| Benchmark | Metric / category | Result |
| :--- | :--- | :--- |
| VINS-4KEval++ | Overall | **8.62**, 2nd of 7 models |
| VINS-4KEval++ | Non-edited-region consistency (NC) | **9.06** |
| ImgEdit-Bench | Hybrid / Action | **4.59 / 4.93**, 1st in both |
| CIE-Bench | Total, before → after ACE-MOPD | **6.24 → 8.87** |

The metadata uses descriptive custom metric identifiers for these measures. See the project page for training details and the broader BluePixel Agent framework.

## Limitations and License

Intended for image editing with English instructions. Outputs may contain artifacts or unintended changes; review results before use.

### Model license

BluePixel-Turbo is based on a modified version of [FLUX.2 [klein] 9B](https://huggingface.co/black-forest-labs/FLUX.2-klein-9B). The model weights are subject to the [FLUX Non-Commercial License](LICENSE.md).

Rights to use the FLUX Model and its derivatives are granted directly by Black Forest Labs Inc. under that license. See [NOTICE](NOTICE) for the required attribution.

For commercial licensing, please refer to [Black Forest Labs licensing](https://bfl.ai/licensing).

### Code license

For the license applicable to the BluePixel code repository, see its [LICENSE](https://github.com/vivo/BluePixel/blob/main/LICENSE) and [NOTICE](https://github.com/vivo/BluePixel/blob/main/NOTICE).