Video-Text-to-Text
Transformers
Safetensors
English
Chinese
mllama
text-generation
multimodal
video
vision-language
sft
custom_code
text-generation-inference
Instructions to use OpenMOSS-Team/moss-video-preview-sft with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use OpenMOSS-Team/moss-video-preview-sft with Transformers:
# Load model directly from transformers import AutoProcessor, AutoModelForCausalLM processor = AutoProcessor.from_pretrained("OpenMOSS-Team/moss-video-preview-sft", trust_remote_code=True) model = AutoModelForCausalLM.from_pretrained("OpenMOSS-Team/moss-video-preview-sft", trust_remote_code=True, device_map="auto") - Notebooks
- Google Colab
- Kaggle
Upload moss-video-sft
Browse files- config.json +2 -2
config.json
CHANGED
|
@@ -3,7 +3,7 @@
|
|
| 3 |
"VideoMllamaForConditionalGeneration"
|
| 4 |
],
|
| 5 |
"image_token_index": 128256,
|
| 6 |
-
"model_type": "
|
| 7 |
"auto_map": {
|
| 8 |
"AutoConfig": "configuration_video_mllama.VideoMllamaConfig",
|
| 9 |
"AutoModelForCausalLM": "modeling_video_mllama.VideoMllamaForConditionalGeneration",
|
|
@@ -228,4 +228,4 @@
|
|
| 228 |
"use_bfloat16": false,
|
| 229 |
"vision_output_dim": 7680
|
| 230 |
}
|
| 231 |
-
}
|
|
|
|
| 3 |
"VideoMllamaForConditionalGeneration"
|
| 4 |
],
|
| 5 |
"image_token_index": 128256,
|
| 6 |
+
"model_type": "mllama",
|
| 7 |
"auto_map": {
|
| 8 |
"AutoConfig": "configuration_video_mllama.VideoMllamaConfig",
|
| 9 |
"AutoModelForCausalLM": "modeling_video_mllama.VideoMllamaForConditionalGeneration",
|
|
|
|
| 228 |
"use_bfloat16": false,
|
| 229 |
"vision_output_dim": 7680
|
| 230 |
}
|
| 231 |
+
}
|