Video-Text-to-Text
Transformers
Safetensors
English
qwen3
text-generation
video-understanding
long-video-understanding
agentic-llm
video-question-answering
vision-language-model
grpo
reinforcement-learning
icml-2026
text-generation-inference
Instructions to use CewEhao/VideoSEAL_8B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use CewEhao/VideoSEAL_8B with Transformers:
# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("CewEhao/VideoSEAL_8B") model = AutoModelForCausalLM.from_pretrained("CewEhao/VideoSEAL_8B") - Notebooks
- Google Colab
- Kaggle
File size: 613 Bytes
14d11e9 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 | {
"additional_special_tokens": [
"<|im_start|>",
"<|im_end|>",
"<|object_ref_start|>",
"<|object_ref_end|>",
"<|box_start|>",
"<|box_end|>",
"<|quad_start|>",
"<|quad_end|>",
"<|vision_start|>",
"<|vision_end|>",
"<|vision_pad|>",
"<|image_pad|>",
"<|video_pad|>"
],
"eos_token": {
"content": "<|im_end|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false
},
"pad_token": {
"content": "<|endoftext|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false
}
}
|