omni-deepseek-v0 / README.md
princepride's picture
Update README.md
c5ec6ca verified
metadata
library_name: transformers
tags: []

Model Card for Omni DeepSeek

Model Training Details

Training Steps

Model Description

I've trained an interesting text-image-video CoT LLM through image-text space alignment, CoT-SFT warm-up, and GRPO training. I've observed that this model can provide more accurate answers through Chain-of-Thought reasoning, particularly excelling in complex instruction following and formatted output.

  • Developed by: princepride
  • Model type: text-image-video CoT LLM

Model Sources

  • Demo :
  • It can analysis the video with chain of thought to output a better result
  • Input :
  • Click to watch the demo
  • Output :
  • Click to watch the demo
  • Because chain of thought, it can better follow complex instructions and generate accurately formatted output:
  • Input :
  • Click to watch the demo
  • Output :
  • Click to watch the demo

Uses(At least 24GB of VRAM is required)

pip install -r requirements.txt

streamlit run app.py