🤗 Hugging Face | 💻 GitHub | 🌐 LightX2V | 💬 WeChat
## Introduction **Four-step image generation and editing. Built with Qwen.** This model is the reinforcement-learning-refined version of LightX2V Lightning v2, distilled from [Qwen-Image-2.1](https://huggingface.co/Qwen/Qwen-Image-2.1). It supports text-to-image generation, single-reference image editing, and multi-reference image generation with **4 denoising steps** and **CFG = 1.0**. RL training was performed with [LightX2V RL](https://github.com/ModelTC/LightX2V/tree/main/lightx2v_rl). ## Model files The release provides the **full transformer weights** in a single `diffusion_pytorch_model.safetensors` file, together with `config.json`. The weights are stored in FP32 and can be loaded in BF16 for inference. `inference_config.json` records the base model, four sampling sigmas, guidance scale, KV cache setting, and scheduler parameters. It is a release-specific configuration file, explicitly read by the example below; Diffusers does not load it automatically. The inference files are: ```text config.json diffusion_pytorch_model.safetensors inference_config.json README.md assets/img_lightx2v.png assets/wechat.jpg assets/comparison/ ``` `merge_manifest.json` is a local export audit record, is not used for inference, and can be excluded when uploading. Load the processor, text encoder, and VAE from the original Qwen-Image-2.1 pipeline, then replace its transformer as shown below. This checkpoint is not a LoRA adapter or a standalone pipeline bundle. ## Quick start Install a recent Diffusers version with `QwenImage21Pipeline` support: ```bash pip install git+https://github.com/huggingface/diffusers.git pip install "transformers>=5.17.0" accelerate safetensors pillow ``` Run the following code from the downloaded model directory, or change `MODEL_DIR` to its local path. The base pipeline components will be downloaded from Hugging Face when needed. ### Load the model ```python import json from pathlib import Path import torch from PIL import Image from diffusers import ( FlowMatchEulerDiscreteScheduler, QwenImage21Pipeline, QwenImage21Transformer2DModel, ) MODEL_DIR = "." with (Path(MODEL_DIR) / "inference_config.json").open(encoding="utf-8") as f: inference_config = json.load(f) SIGMAS = inference_config["sample_sigmas"] transformer = QwenImage21Transformer2DModel.from_pretrained( MODEL_DIR, torch_dtype=torch.bfloat16, ) pipe = QwenImage21Pipeline.from_pretrained( inference_config["base_model"], transformer=transformer, torch_dtype=torch.bfloat16, ) # The four sigmas are already shifted. Do not apply dynamic shifting again. pipe.scheduler = FlowMatchEulerDiscreteScheduler( **inference_config["scheduler"], ) pipe.to("cuda") def generate(prompt, image=None, seed=42): return pipe( prompt=prompt, image=image, width=1024, height=1024, sigmas=SIGMAS, num_inference_steps=len(SIGMAS), true_cfg_scale=inference_config["true_cfg_scale"], use_kv_cache=inference_config["use_kv_cache"], generator=torch.Generator("cpu").manual_seed(seed), ).images[0] ``` Use this exact sigma schedule for all three tasks. The terminal sigma of zero is appended by the scheduler. Setting only `num_inference_steps=4` does not reproduce this schedule. ### Text-to-image ```python image = generate( "A ceramic teapot beside a small bowl of oranges on a wooden table, " "soft morning window light, detailed glaze, natural colors." ) image.save("t2i.png") ``` ### Single-reference editing ```python reference = Image.open("reference.png").convert("RGB") image = generate( "Change the jacket to dark green. Preserve the person's face, pose, " "and the rest of the scene.", image=[reference], ) image.save("edit.png") ``` ### Multi-reference generation Pass reference images in the same order used by the prompt: ```python references = [ Image.open("person.png").convert("RGB"), Image.open("jacket.png").convert("RGB"), ] image = generate( "Show the person from image 1 wearing the jacket from image 2. " "Preserve the person's identity and the jacket's design, " "with natural lighting and a simple studio background.", image=references, ) image.save("multi_reference.png") ``` The examples use 1024 × 1024 output. For other aspect ratios, set both dimensions to multiples of 32. Generation quality can vary with the prompt and seed; fine text, anatomy, object counts, and precise reference preservation may still require checking. ## Comparison with Qwen-Image-2.1 (40 steps) Selected qualitative examples from our validation runs: **Lightning v2 RL uses 4 steps**, while the baseline uses the original [Qwen-Image-2.1](https://huggingface.co/Qwen/Qwen-Image-2.1) weights with 40-step Euler sampling. Both use CFG = 1.0, the same full enhanced prompts, the same input references where applicable, and 1024 × 1024 output. ### Text-to-image | Example / prompt | Qwen-Image-2.1 · 40 steps | Lightning v2 RL · 4 steps | | --- | --- | --- | | [Dolphin at a Burning Shoreline](https://huggingface.co/lightx2v/Qwen-Image-2.1-Lightning-v2-RL/blob/main/assets/comparison/t2i/12/prompt.txt) |
| 
|
| [Weathered Bicycle in the Rain](https://huggingface.co/lightx2v/Qwen-Image-2.1-Lightning-v2-RL/blob/main/assets/comparison/t2i/13/prompt.txt) | 
| 
|
### Single-reference editing
| Example / prompt | Reference images | Qwen-Image-2.1 · 40 steps | Lightning v2 RL · 4 steps |
| --- | --- | --- | --- |
| [Flying Dragon above Waterfall Gardens](https://huggingface.co/lightx2v/Qwen-Image-2.1-Lightning-v2-RL/blob/main/assets/comparison/i2i_single/20/prompt.txt) | 
| 
| 
|
| [Guitarist and Cat in a Busy Marketplace](https://huggingface.co/lightx2v/Qwen-Image-2.1-Lightning-v2-RL/blob/main/assets/comparison/i2i_single/22/prompt.txt) | 
| 
| 
|
### Multi-reference generation
| Example / prompt | Reference images | Qwen-Image-2.1 · 40 steps | Lightning v2 RL · 4 steps |
| --- | --- | --- | --- |
| [Food Transfer into a Stylized Scene](https://huggingface.co/lightx2v/Qwen-Image-2.1-Lightning-v2-RL/blob/main/assets/comparison/i2i_multi/27/prompt.txt) | 
| 
| 
|
| [Floral Porcelain Teapot in a Portrait Scene](https://huggingface.co/lightx2v/Qwen-Image-2.1-Lightning-v2-RL/blob/main/assets/comparison/i2i_multi/29/prompt.txt) | 
| 
| 
|
## License and acknowledgements
This model is derived from Qwen-Image-2.1 and is subject to the [Qwen Research License](https://huggingface.co/Qwen/Qwen-Image-2.1/blob/main/LICENSE), including its non-commercial terms.
Thanks to the Qwen team and the Diffusers community. See [LightX2V](https://github.com/ModelTC/LightX2V), the earlier [Qwen-Image-2.1-Lightning](https://huggingface.co/lightx2v/Qwen-Image-2.1-Lightning) release, and [Qwen-Image-2.1-Turbo](https://huggingface.co/Qwen/Qwen-Image-2.1-Turbo) for related work.
