VideoChat TPO
This model is based on the paper Task Preference Optimization: Improving Multimodal Large Language Models with Vision Task Alignment.
Model source
- Kind
- Other
- File size
- 1.23–4.65 GB
- Base model
- mistralai/Mistral-7B-Instruct-v0.2
- Version
- vRevision 7d22760
- Updated
- Oct 1, 2026
- Downloads
- 27
- Source
- Hugging Face
Source description
VideoChat2-TPO
This model is based on the paper Task Preference Optimization: Improving Multimodal Large Language Models with Vision Task Alignment.
🏃 Installation
pip install -r requirements.txt
python app.py
🔧 Usage
from transformers import AutoModel, AutoTokenizer
from tokenizer import MultimodalLlamaTokenizer
model_path = "OpenGVLab/VideoChat-TPO"
tokenizer = AutoTokenizer.from_pretrained(model_path,
trust_remote_code=True,
use_fast=False,)
model = AutoModel.from_pretrained(model_path, trust_remote_code=True, _tokenizer=self.tokenizer).eval()
Sources
1 source
Sources
1 sourceVerified Aug 22
Model artifacts
4 artifacts
Model artifacts
4 artifactsmodel-00001-of-00004.safetensors
safetensors · 4.65 GB · SHA-256 c8640081ef34…a330 · Hugging Face
Downloadmodel-00002-of-00004.safetensors
safetensors · 4.61 GB · SHA-256 8907fc1447bb…086c · Hugging Face
Downloadmodel-00003-of-00004.safetensors
safetensors · 4.61 GB · SHA-256 5640bcb11fc2…4f93 · Hugging Face
Downloadmodel-00004-of-00004.safetensors
safetensors · 1.23 GB · SHA-256 d0360c7932a2…4fe8 · Hugging Face
DownloadSource excerpts
2 excerpts
Source excerpts
2 excerpts--- base_model: - mistralai/Mistral-7B-Instruct-v0.2 library_name: transformers license: mit pipeline_tag: video-text-to-text --- # VideoChat2-TPO This model is based on the paper [Task Preference Optimization: Improving Multimodal Large Language Models with Vision Task Alignment](https://huggingface.co/papers/2412.19326). ## 🏃 Installation ``` pip install -r requirements.txt python app.py ``` ## 🔧 Usage ``` from transformers import AutoModel, AutoTokenizer from tokenizer import MultimodalLlamaTokenizer model_path = "OpenGVLab/VideoChat-TPO" tokenizer = AutoTokenizer.from_pretrained(model_path, trust_remote_code=True, use_fast=False,) model = AutoModel.from_pretrained(model_path, trust_remote_code=True, _tokenizer=self.tokenizer).eval() ```
Source context: 27 downloads · 5 likes · Pipeline video-text-to-text · Library transformers · Repo OpenGVLab/VideoChat-TPO
