About
Qwen 2.5 Vision-Language 7B text/vision encoder for Qwen Image Edit.
Handles prompt processing and image understanding for the Qwen Image editing pipeline.
Available in full bf16 (16.6 GB) and fp8 quantized (9.4 GB) variants.
Variants
| Variant | Format | Size | VRAM | Install |
|---|---|---|---|---|
| fp8 ⓘ | safetensors | 10.1 GB | 10+ GB | mods install qwen-image-clip --variant fp8 |
| bf16 ⓘ | safetensors | 17.8 GB | 16+ GB | mods install qwen-image-clip --variant bf16 |