YAML Metadata Warning:empty or missing yaml metadata in repo card

Check out the documentation for more information.

base_model: Qwen/Qwen-Image-Edit-2511 library_name: diffusers license: apache-2.0 tags: - qwen-image-edit - lora - multi-angle - camera-angles - image-to-image - quantized - nf4

Qwen-Image-Edit-2511 + Multiple Angles LoRA (Full NF4 โ€” Transformer + Text Encoder)

Merged and fully quantized version of Qwen/Qwen-Image-Edit-2511 with fal/Qwen-Image-Edit-2511-Multiple-Angles-LoRA fused into the weights.

Both the DiT transformer and the Qwen2.5-VL text encoder are quantized to NF4 4-bit with double quantization. Runs on ~15GB VRAM โ€” fits a T4.

Usage

import torch
from diffusers import QwenImageEditPlusPipeline, QwenImageTransformer2DModel
from transformers import Qwen2_5_VLForConditionalGeneration, BitsAndBytesConfig
from PIL import Image

quant_config = BitsAndBytesConfig(
    load_in_4bit=True,
    bnb_4bit_quant_type="nf4",
    bnb_4bit_compute_dtype=torch.bfloat16,
    bnb_4bit_use_double_quant=True,
)

transformer = QwenImageTransformer2DModel.from_pretrained(
    "mash2005/Qwen-Image-Edit-2511-MultiAngles-FullQ4",
    subfolder="transformer",
    torch_dtype=torch.bfloat16,
    device_map="cuda"
)

text_encoder = Qwen2_5_VLForConditionalGeneration.from_pretrained(
    "mash2005/Qwen-Image-Edit-2511-MultiAngles-FullQ4",
    subfolder="text_encoder",
    torch_dtype=torch.bfloat16,
    device_map="cuda"
)

pipeline = QwenImageEditPlusPipeline.from_pretrained(
    "mash2005/Qwen-Image-Edit-2511-MultiAngles-FullQ4",
    transformer=transformer,
    text_encoder=text_encoder,
    torch_dtype=torch.bfloat16,
)
pipeline.enable_model_cpu_offload()

image = Image.open("your_image.jpg").convert("RGB")

result = pipeline(
    image=[image],
    prompt="Show this object from the back view, azimuth 180 degrees",
    negative_prompt=" ",
    num_inference_steps=30,
    guidance_scale=1.0,
    true_cfg_scale=4.0,
).images[0]

result.save("output.png")

Prompt format

  • "front view, azimuth 0 degrees"
  • "back view, azimuth 180 degrees"
  • "side view, azimuth 90 degrees"
  • "top-down view, elevation 90 degrees"

Credits

Downloads last month
31
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support