Qwen-Image
Collection
Contains FP8, INT8, INT4 and FP4 weight-only safetensors quants of the Qwen-Image series. • 4 items • Updated
Text-to-Image
Diffusers
Safetensors
QwenImage21Pipeline
quantized
mecha
weight-only
int4
experimental
image
8-bit precision
import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("Rin247/Qwen-Image-2.1-INT4", dtype=torch.bfloat16, device_map="cuda")
prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k"
image = pipe(prompt).images[0]Qwen-Image-2.1-INT4
INT4 safetensors media package of Qwen-Image-2.1.
MECHA Forge
MECHA — Model Engineering & Contextual Hearth Assembly
This media package has been forged and optimized in the MECHA Forge. It includes the Denoiser, Text Encoder, and VAE components packaged together for easier deployment.
Quantization Details
- Base model:
Qwen-Image-2.1 - Format: INT4
- Components: Denoiser, Text Encoder, VAE
- VAE: protected from 4-bit; uses INT8 when a 4-bit format is selected
- Downloads last month
- 72