Implementation:Open compass VLMEvalKit Emu

Field	Value
source	VLMEvalKit
domain	Vision, Model_Architecture

Overview

VLM adapter for the Emu2-Chat model enabling benchmark evaluation in VLMEvalKit.

Description

Emu inherits from BaseModel and wraps the Emu2-Chat model for use within the VLMEvalKit evaluation framework. It initializes the model and tokenizer/processor from a HuggingFace model path (default: BAAI/Emu2-Chat) and provides the generate_inner method for inference. Also includes Emu3_chat and Emu3_gen adapter classes for Emu3 variants.

Usage

Register in vlmeval/config.py via supported_VLM and invoke through the standard evaluation pipeline.

Code Reference

Source: vlmeval/vlm/emu.py, Lines: L1-304
Import: from vlmeval.vlm.emu import Emu

Signature:

class Emu(BaseModel):
    INSTALL_REQ = False
    INTERLEAVE = True
    def __init__(self, model_path='BAAI/Emu2-Chat', **kwargs): ...
    def generate_inner(self, message, dataset=None): ...

I/O Contract

Direction	Description
Inputs	message — list of dicts with type (text/image) and value; dataset — optional dataset name for custom prompting
Outputs	generate_inner() returns str (model response text)

Usage Examples

from vlmeval.vlm.emu import Emu
model = Emu(model_path='path/to/model')
response = model.generate_inner(message)

Related Pages

Page Connections

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

Principle

Implementation

Heuristic

Environment