Image-Text-to-Text
Transformers
qwen3_5
embedding
multimodal
quantized
fp8
video-text-to-text
conversational
custom_code
Instructions to use Weidows/WeMM-Embedding-2B-FP8 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Weidows/WeMM-Embedding-2B-FP8 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-text-to-text", model="Weidows/WeMM-Embedding-2B-FP8", trust_remote_code=True) messages = [ { "role": "user", "content": [ {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"}, {"type": "text", "text": "What animal is on the candy?"} ] }, ] pipe(text=messages)# pip install -U transformers accelerate # Load model directly from transformers import AutoProcessor, AutoModelForMultimodalLM processor = AutoProcessor.from_pretrained("Weidows/WeMM-Embedding-2B-FP8", trust_remote_code=True) model = AutoModelForMultimodalLM.from_pretrained("Weidows/WeMM-Embedding-2B-FP8", trust_remote_code=True, device_map="auto") messages = [ { "role": "user", "content": [ {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"}, {"type": "text", "text": "What animal is on the candy?"} ] }, ] inputs = processor.apply_chat_template( messages, add_generation_prompt=True, tokenize=True, return_dict=True, return_tensors="pt", ).to(model.device) outputs = model.generate(**inputs, max_new_tokens=256) print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:])) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use Weidows/WeMM-Embedding-2B-FP8 with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "Weidows/WeMM-Embedding-2B-FP8" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Weidows/WeMM-Embedding-2B-FP8", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Describe this image in one sentence." }, { "type": "image_url", "image_url": { "url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg" } } ] } ] }'Use Docker
docker model run hf.co/Weidows/WeMM-Embedding-2B-FP8
- SGLang
How to use Weidows/WeMM-Embedding-2B-FP8 with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "Weidows/WeMM-Embedding-2B-FP8" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Weidows/WeMM-Embedding-2B-FP8", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Describe this image in one sentence." }, { "type": "image_url", "image_url": { "url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg" } } ] } ] }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "Weidows/WeMM-Embedding-2B-FP8" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Weidows/WeMM-Embedding-2B-FP8", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Describe this image in one sentence." }, { "type": "image_url", "image_url": { "url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg" } } ] } ] }' - Docker Model Runner
How to use Weidows/WeMM-Embedding-2B-FP8 with Docker Model Runner:
docker model run hf.co/Weidows/WeMM-Embedding-2B-FP8
Download patch_sglang_video.py from Weidows/WeMM-Embedding-2B-FP8: direct link, hf CLI and curl.
- Browser
- Download file 1.98 kB
-
https://huggingface.co/Weidows/WeMM-Embedding-2B-FP8/resolve/main/patch_sglang_video.py
- Command line
-
hf download hf://Weidows/WeMM-Embedding-2B-FP8/patch_sglang_video.py
-
curl -L -o patch_sglang_video.py https://huggingface.co/Weidows/WeMM-Embedding-2B-FP8/resolve/main/patch_sglang_video.py
1.98 kB
| #!/usr/bin/env python3 | |
| """Align SGLang 0.5.9 video preprocessing with WeMM-Embedding.""" | |
| import importlib.metadata | |
| import py_compile | |
| import shutil | |
| from pathlib import Path | |
| import sglang | |
| EXPECTED_VERSION = "0.5.9" | |
| REPLACEMENTS = ( | |
| ( | |
| "IMAGE_FACTOR = 28", | |
| "IMAGE_FACTOR = 32 # patch_size=16 * spatial_merge_size=2", | |
| ), | |
| ( | |
| " idx = np.linspace(0, total_frames - 1, num=nframes, dtype=np.int64)", | |
| " idx = torch.linspace(0, total_frames - 1, nframes).round().long().cpu().numpy()", | |
| ), | |
| ( | |
| """ video = torchvision.transforms.functional.resize( | |
| video, | |
| [resized_height, resized_width], | |
| interpolation=InterpolationMode.BILINEAR, | |
| ) | |
| """, | |
| """ video = torchvision.transforms.functional.resize( | |
| video, | |
| [resized_height, resized_width], | |
| interpolation=InterpolationMode.BICUBIC, | |
| antialias=True, | |
| ).float() | |
| """, | |
| ), | |
| ) | |
| def main() -> None: | |
| version = importlib.metadata.version("sglang") | |
| if version != EXPECTED_VERSION: | |
| raise RuntimeError(f"Expected sglang=={EXPECTED_VERSION}, found {version}") | |
| path = ( | |
| Path(sglang.__file__).resolve().parent | |
| / "srt" | |
| / "multimodal" | |
| / "processors" | |
| / "qwen_vl.py" | |
| ) | |
| text = path.read_text(encoding="utf-8") | |
| if all(new in text for _, new in REPLACEMENTS): | |
| print("SGLang video preprocessing is ready.") | |
| return | |
| updated = text | |
| for old, new in REPLACEMENTS: | |
| if updated.count(old) != 1: | |
| raise RuntimeError(f"Unexpected SGLang source: {old.splitlines()[0]}") | |
| updated = updated.replace(old, new, 1) | |
| backup = path.with_suffix(path.suffix + ".original") | |
| if not backup.exists(): | |
| shutil.copy2(path, backup) | |
| path.write_text(updated, encoding="utf-8") | |
| py_compile.compile(str(path), doraise=True) | |
| print("SGLang video preprocessing is ready.") | |
| if __name__ == "__main__": | |
| main() | |