| --- |
| base_model: Qwen/Qwen-Image |
| library_name: diffusers |
| license: apache-2.0 |
| pipeline_tag: text-to-image |
| tags: |
| - qwen-image |
| - text-to-image |
| - quantized |
| - fp8 |
| - e4m3 |
| - diffusers |
| base_model_relation: quantized |
| --- |
| # Qwen-Image-e4m3 |
|
|
| FP8 (e4m3) dynamically-quantized [Qwen-Image](https://huggingface.co/Qwen/Qwen-Image), |
| saved as a complete `QwenImagePipeline`. |
|
|
| ## What was changed |
|
|
| All 60 blocks of the `QwenImageTransformer2DModel` are quantized to |
| `e4m3_e4m3_dynamic` — `float8_e4m3fn` weights with dynamically scaled |
| `float8_e4m3fn` activations. The Qwen2.5-VL text encoder, the VAE and the |
| transformer's non-block tensors are untouched and stay in bf16. The transformer |
| drops from ~40.9 GB to ~20.5 GB. |
|
|
| Quantization was done with WaveSpeed's `xelerate.ao.quantize`. Weights are |
| stored as pickled `.bin` shards, so loading requires `use_safetensors=False`. |
|
|
| FP8 matmul needs Hopper (H100/H200) or newer to be faster than bf16; on older |
| GPUs the weights are dequantized on the fly and you only get the memory saving. |
|
|
| One caveat worth repeating from our own testing: on Qwen-Image, running fp8 |
| *weight-and-activation* matmul under a fully fused fast path produces visible |
| quality loss. The configuration published here — dynamic per-tensor activation |
| scaling with bf16 accumulation — is the one that holds up. |
|
|
| ## Usage |
|
|
| ```python |
| import torch |
| from diffusers import QwenImagePipeline |
| |
| pipe = QwenImagePipeline.from_pretrained( |
| "wavespeed/Qwen-Image-e4m3", |
| torch_dtype=torch.bfloat16, |
| use_safetensors=False, |
| ).to("cuda") |
| |
| image = pipe("a chalkboard menu written in neat cursive").images[0] |
| ``` |
|
|
| ## License |
|
|
| Apache-2.0, inherited from Qwen-Image. |
|
|
