multimodalart HF Staff commited on
Commit
7b97603
·
verified ·
1 Parent(s): 04034e3

Delete tmp/hugging-demos-build-model_OpenMOSS-Team_MOSS-VL-Instruct-0708-t4yujp0c/build/requirements.txt with huggingface_hub

Browse files
tmp/hugging-demos-build-model_OpenMOSS-Team_MOSS-VL-Instruct-0708-t4yujp0c/build/requirements.txt DELETED
@@ -1,26 +0,0 @@
1
- ---
2
- title: MOSS-VL-Instruct-0708
3
- emoji: 🧠
4
- colorFrom: gray
5
- colorTo: blue
6
- sdk: gradio
7
- sdk_version: 5.50.0
8
- app_file: app.py
9
- short_description: Image and video understanding with MOSS-VL multimodal model
10
- python_version: "3.12"
11
- startup_duration_timeout: 1h
12
- ---
13
-
14
- # MOSS-VL-Instruct-0708
15
-
16
- An 11B parameter vision-language model from OpenMOSS that supports both image and video understanding.
17
-
18
- ## Capabilities
19
-
20
- - **Image understanding**: OCR, document parsing, fine-grained visual recognition, multi-image comparison
21
- - **Video understanding**: Long-form video comprehension, temporal reasoning, action recognition
22
- - **256K context window** for processing long videos and complex instructions
23
-
24
- ## Usage
25
-
26
- Upload an image or video and enter a text prompt describing what you want the model to do. The model will generate a text response based on its understanding of the visual input.