config/chatbot_ui.yaml

WORK_DIR: ./cache/chatbot
FILE_SYSTEM:
  - NAME: LocalFs
    TEMP_DIR: ./cache
  - NAME: HttpFs
    TEMP_DIR: ./cache
  - NAME: HuggingfaceFs
    TEMP_DIR: ./cache
  - NAME: ModelscopeFs
    TEMP_DIR: ./cache
#
ENABLE_I2V: False
#
MODEL:
  EDIT_MODEL:
    MODEL_CFG_DIR: config/models/
  I2V:
    MODEL_NAME: CogVideoX-5b-I2V
    MODEL_DIR: ms://ZhipuAI/CogVideoX-5b-I2V/
  CAPTIONER:
    MODEL_NAME: InternVL2-2B
    MODEL_DIR: ms://OpenGVLab/InternVL2-2B/
    PROMPT: '<image>\nThis image is the first frame of a video. Based on this image, please imagine what changes may occur in the next few seconds of the video. Please output brief description, such as "a dog running" or "a person turns to left". No more than 30 words.'
  ENHANCER:
    MODEL_NAME: Meta-Llama-3.1-8B-Instruct
    MODEL_DIR: ms://LLM-Research/Meta-Llama-3.1-8B-Instruct/