Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,18 @@
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
| 3 |
+
language:
|
| 4 |
+
- en
|
| 5 |
+
- zh
|
| 6 |
+
base_model: Qwen/Qwen-Image-Layered
|
| 7 |
+
base_model_relation: quantized
|
| 8 |
+
pipeline_tag: image-text-to-image
|
| 9 |
+
library_name: diffusers
|
| 10 |
---
|
| 11 |
+
# Qwen-Image-Layered (FP8 E5M2 & E4M3FN)
|
| 12 |
+
|
| 13 |
+
This is a quantization of [Qwen/Qwen-Image-Layered](https://huggingface.co/Qwen/Qwen-Image-Layered) to **FP8 E5M2** and **FP8 E4M3FN**.
|
| 14 |
+
|
| 15 |
+
Sensitive layers (norms, embeddings, biases) were kept in BF16.
|
| 16 |
+
|
| 17 |
+
**License & Usage:**
|
| 18 |
+
This model strictly follows the original licensing terms and usage restrictions. Please refer to the [original model card](https://huggingface.co/Qwen/Qwen-Image-Layered) for details.
|