Image-Text-to-Text
Transformers
Safetensors
qwen3_vl
qwen3-vl
viss
streaming-video
visual-question-answering
causal-video
conversational
Instructions to use mis2/ViSS-8B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use mis2/ViSS-8B with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-text-to-text", model="mis2/ViSS-8B") 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("mis2/ViSS-8B") model = AutoModelForMultimodalLM.from_pretrained("mis2/ViSS-8B", 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 mis2/ViSS-8B with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "mis2/ViSS-8B" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "mis2/ViSS-8B", "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/mis2/ViSS-8B
- SGLang
How to use mis2/ViSS-8B 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 "mis2/ViSS-8B" \ --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": "mis2/ViSS-8B", "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 "mis2/ViSS-8B" \ --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": "mis2/ViSS-8B", "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 mis2/ViSS-8B with Docker Model Runner:
docker model run hf.co/mis2/ViSS-8B
Download code/viss_core/data_v0/trainer_adapter.py from mis2/ViSS-8B: direct link, hf CLI and curl.
- Browser
- Download file 3.27 kB
-
https://huggingface.co/mis2/ViSS-8B/resolve/main/code/viss_core/data_v0/trainer_adapter.py
- Command line
-
hf download hf://mis2/ViSS-8B/code/viss_core/data_v0/trainer_adapter.py
-
curl -L -o trainer_adapter.py https://huggingface.co/mis2/ViSS-8B/resolve/main/code/viss_core/data_v0/trainer_adapter.py
3.27 kB
| """Strict adapter from a sealed semantic release to Qwen messages.""" | |
| from __future__ import annotations | |
| from pathlib import Path | |
| from typing import Any | |
| from PIL import Image | |
| from viss_core.data_v0.semantic_release import ( | |
| infer_semantic_release_root, | |
| load_semantic_sft_rows, | |
| ) | |
| def load_training_rows( | |
| path: Path, *, allow_nontrainable: bool = False | |
| ) -> list[dict[str, Any]]: | |
| """Load only a validated, training-authorized semantic release.""" | |
| if allow_nontrainable: | |
| raise RuntimeError("the release trainer has no non-trainable bypass") | |
| path = path.resolve() | |
| if path.suffix != ".jsonl": | |
| raise ValueError("semantic training accepts canonical JSONL only") | |
| return load_semantic_sft_rows(path) | |
| def infer_release_root(path: Path) -> Path | None: | |
| return infer_semantic_release_root(path) | |
| def normalize_messages( | |
| messages: Any, | |
| *, | |
| media_root: Path | None = None, | |
| ) -> list[dict[str, Any]]: | |
| """Resolve release-local images and preserve assistant-only loss ownership.""" | |
| if not isinstance(messages, list): | |
| raise TypeError("semantic messages must be a list") | |
| roles = [message.get("role") for message in messages] | |
| if roles != ["user", "assistant"]: | |
| raise ValueError("semantic messages require user then assistant") | |
| root = media_root.resolve() if media_root is not None else None | |
| if root is None: | |
| raise RuntimeError("semantic media requires an explicit release root") | |
| output: list[dict[str, Any]] = [] | |
| for message in messages: | |
| raw_content = message.get("content") | |
| if not isinstance(raw_content, list) or not raw_content: | |
| raise TypeError("semantic message content must be a non-empty list") | |
| content: list[dict[str, Any]] = [] | |
| for item in raw_content: | |
| if not isinstance(item, dict): | |
| raise TypeError("semantic content item must be an object") | |
| if item.get("type") == "text": | |
| content.append({"type": "text", "text": str(item.get("text") or "")}) | |
| continue | |
| if item.get("type") != "image" or not isinstance(item.get("image"), str): | |
| raise ValueError("semantic content item has an unsupported type") | |
| if message["role"] != "user": | |
| raise ValueError("only user states may contain semantic images") | |
| relative = Path(item["image"]) | |
| if relative.is_absolute() or ".." in relative.parts: | |
| raise RuntimeError("semantic image path must be release-relative") | |
| image_path = (root / relative).resolve(strict=True) | |
| try: | |
| image_path.relative_to(root) | |
| except ValueError as error: | |
| raise RuntimeError("semantic image path escapes the release") from error | |
| if not image_path.is_file() or image_path.is_symlink(): | |
| raise RuntimeError("semantic image is missing or unsafe") | |
| content.append({"type": "image", "image": Image.open(image_path).convert("RGB")}) | |
| output.append( | |
| { | |
| "role": str(message["role"]), | |
| "content": content, | |
| "loss": bool(message.get("loss")), | |
| } | |
| ) | |
| return output | |