Video-Text-to-Text
Transformers
Safetensors
English
Chinese
videochat3
feature-extraction
video-language-model
vision-language-model
multimodal
video-understanding
image-understanding
streaming-video
custom_code
Instructions to use MCG-NJU/VideoChat3-4B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use MCG-NJU/VideoChat3-4B with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("MCG-NJU/VideoChat3-4B", trust_remote_code=True, dtype="auto") - Notebooks
- Google Colab
- Kaggle
| { | |
| "architectures": [ | |
| "VideoChat3ForConditionalGeneration" | |
| ], | |
| "auto_map": { | |
| "AutoProcessor": "processing_videochat3.VideoChat3Processor", | |
| "AutoConfig": "configuration_videochat3.VideoChat3Config", | |
| "AutoModel": "modeling_videochat3.VideoChat3ForConditionalGeneration", | |
| "AutoModelForCausalLM": "modeling_videochat3.VideoChat3ForConditionalGeneration" | |
| }, | |
| "image_token_id": 151655, | |
| "model_type": "videochat3", | |
| "text_config": { | |
| "architectures": [ | |
| "Qwen3ForCausalLM" | |
| ], | |
| "attention_bias": false, | |
| "attention_dropout": 0.0, | |
| "bos_token_id": 151643, | |
| "eos_token_id": 151645, | |
| "head_dim": 128, | |
| "hidden_act": "silu", | |
| "hidden_size": 2560, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 9728, | |
| "max_position_embeddings": 262144, | |
| "max_window_layers": 36, | |
| "model_type": "qwen3", | |
| "num_attention_heads": 32, | |
| "num_hidden_layers": 36, | |
| "num_key_value_heads": 8, | |
| "rms_norm_eps": 1e-06, | |
| "rope_scaling": null, | |
| "rope_theta": 5000000, | |
| "sliding_window": null, | |
| "tie_word_embeddings": true, | |
| "dtype": "bfloat16", | |
| "use_cache": true, | |
| "use_sliding_window": false, | |
| "vocab_size": 151936 | |
| }, | |
| "tie_word_embeddings": false, | |
| "transformers_version": "4.57.0.dev0", | |
| "video_token_id": 151656, | |
| "vision_config": { | |
| "patch_size": 14, | |
| "num_attention_heads": 16, | |
| "num_hidden_layers": 27, | |
| "hidden_size": 1152, | |
| "intermediate_size": 4304, | |
| "init_pos_emb_height": 64, | |
| "init_pos_emb_width": 64, | |
| "merge_kernel_size": [ | |
| 2, | |
| 2 | |
| ], | |
| "temporal_patch_size": 1, | |
| "temporal_merge_size": 4, | |
| "dtype": "bfloat16", | |
| "attn_impl": "flash_attention_2" | |
| }, | |
| "vision_end_token_id": 151653, | |
| "vision_start_token_id": 151652 | |
| } | |