Implementation:Open compass VLMEvalKit VideoChatOnlineV2Wrapper

Field	Value
source	VLMEvalKit
domain	Vision, API_Integration

Overview

VideoChatOnlineV2Wrapper provides a VLMEvalKit API adapter for the VideoChat Online V2 vision-language model.

Description

VideoChatOnlineV2Wrapper inherits from BaseAPI and communicates with a configurable API endpoint for the VideoChat Online V2 model. It supports custom prompt construction for MCQ and VQA dataset types (excluding MMMU), handles image dumping and base64 encoding, and includes configurable URL and key parameters. The INTERLEAVE flag is set to False, indicating sequential image-text processing.

Usage

Use this adapter when evaluating VideoChat Online V2 models through their API endpoint.

Code Reference

Source: vlmeval/api/video_chat_online_v2.py, Lines: L1-282
Import: from vlmeval.api.video_chat_online_v2 import VideoChatOnlineV2Wrapper

Signature:

class VideoChatOnlineV2Wrapper(BaseAPI):
    def __init__(self, model='VideoChatOnlineV2', retry=5, wait=5,
                 url='', key='', verbose=True, system_prompt=None,
                 temperature=0.7, max_tokens=2048, proxy=None,
                 **kwargs): ...
    def generate_inner(self, inputs, **kwargs): ...

I/O Contract

Direction	Description
Inputs	message — text/image/video content list; model-specific params via kwargs
Outputs	generate() returns str prediction; generate_inner() returns (int, str, str) tuple

Usage Examples

# Example instantiation
model = VideoChatOnlineV2Wrapper(model='VideoChatOnlineV2')
response = model.generate(message)

Related Pages

Page Connections

Double-click a node to navigate. Hold to expand connections.

Principle

Implementation

Heuristic

Environment