Implementation:Open compass VLMEvalKit RBDash

Field	Value
source	VLMEvalKit
domain	Vision, Model_Architecture

Overview

VLM adapter for the RBDash model enabling benchmark evaluation in VLMEvalKit.

Description

RBDash inherits from BaseModel and wraps the RBDash model for use within the VLMEvalKit evaluation framework. It initializes the model and tokenizer/processor from a HuggingFace model path (default: RBDash-Team/RBDash-v1.5) and provides the generate_inner method for inference. Requires a separate code directory root parameter and downloads additional vision encoders during initialization.

Usage

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

Code Reference

Source: vlmeval/vlm/rbdash.py, Lines: L1-282
Import: from vlmeval.vlm.rbdash import RBDash

Signature:

class RBDash(BaseModel):
    INSTALL_REQ = True
    INTERLEAVE = False
    def __init__(self, model_path='RBDash-Team/RBDash-v1.5', **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.rbdash import RBDash
model = RBDash(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