---
tags:
- lora
widget:
- text: >-
    f"<|begin_of_text|><|start_header_id|>system<|end_header_id|>
    You are an expert in financial misinformation
    detection.<|eot_id|>
    <|start_header_id|>user<|end_header_id|>
    {model_input}
    image information:
    {data['image_info']}<|eot_id|><|start_header_id|>assistant<|end_header_id|>"
  parameters:
    negative_prompt: >-
      f"<|begin_of_text|><|start_header_id|>system<|end_header_id|>
      You are an expert in financial misinformation detection.
      <|eot_id|><|start_header_id|>user<|end_header_id|>
      {model_input}
      image information:
      {data['image_info']}<|eot_id|><|start_header_id|>assistant<|end_header_id|>"
  output:
    url: images/86c33077b9aba20177a61a561601069.png
base_model: meta-llama/Llama-3.2-11B-Vision-Instruct
instance_prompt: null
license: llama3.2
---
# FMDMllama

<Gallery />

## Model description 

The model is based on Llama-3.2-11B-Vision-Instruct to fine-tuned.

The lora_weight.zip is the lora model.

The test-datasets.zip is the test datasets.

The inference.py is the inference.

## Download model


[Download](/LuozzzzzzzzzzzzzzY/FMDMllama/tree/main) them in the Files & versions tab.