Video-Text-to-Text
Transformers
Safetensors
English
Chinese
mllama
text-generation
multimodal
video
vision-language
custom_code
text-generation-inference
Instructions to use OpenMOSS-Team/moss-video-preview-base with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use OpenMOSS-Team/moss-video-preview-base with Transformers:
# Load model directly from transformers import AutoProcessor, AutoModelForCausalLM processor = AutoProcessor.from_pretrained("OpenMOSS-Team/moss-video-preview-base", trust_remote_code=True) model = AutoModelForCausalLM.from_pretrained("OpenMOSS-Team/moss-video-preview-base", trust_remote_code=True, device_map="auto") - Notebooks
- Google Colab
- Kaggle
Upload moss-video-preview-base
Browse files- config.json +1 -1
config.json
CHANGED
|
@@ -2,7 +2,7 @@
|
|
| 2 |
"architectures": [
|
| 3 |
"VideoMllamaForConditionalGeneration"
|
| 4 |
],
|
| 5 |
-
"model_type": "
|
| 6 |
"auto_map": {
|
| 7 |
"AutoConfig": "configuration_video_mllama.VideoMllamaConfig",
|
| 8 |
"AutoModelForCausalLM": "modeling_video_mllama.VideoMllamaForConditionalGeneration",
|
|
|
|
| 2 |
"architectures": [
|
| 3 |
"VideoMllamaForConditionalGeneration"
|
| 4 |
],
|
| 5 |
+
"model_type": "mllama",
|
| 6 |
"auto_map": {
|
| 7 |
"AutoConfig": "configuration_video_mllama.VideoMllamaConfig",
|
| 8 |
"AutoModelForCausalLM": "modeling_video_mllama.VideoMllamaForConditionalGeneration",
|