Implementation:Open compass VLMEvalKit llama vision
Appearance
| Field | Value |
|---|---|
| source | VLMEvalKit |
| domain | Vision, Model_Architecture |
Overview
VLM adapter for the Llama-3.2-Vision model enabling benchmark evaluation in VLMEvalKit.
Description
llama_vision inherits from BaseModel and wraps the Llama-3.2-Vision model for use within the VLMEvalKit evaluation framework. It initializes the model and tokenizer/processor from a HuggingFace model path (default: meta-llama/Llama-3.2-11B-Vision-Instruct) and provides the generate_inner method for inference.
Usage
Register in vlmeval/config.py via supported_VLM and invoke through the standard evaluation pipeline.
Code Reference
- Source:
vlmeval/vlm/llama_vision.py, Lines: L1-153 - Import:
from vlmeval.vlm.llama_vision import llama_vision
Signature:
class llama_vision(BaseModel):
INSTALL_REQ = False
INTERLEAVE = True
def __init__(self, model_path='meta-llama/Llama-3.2-11B-Vision-Instruct', **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.llama_vision import llama_vision
model = llama_vision(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