# BLIP2-Clintox-Finetuned
This model is a fine-tuned version of [Salesforce/blip-vqa-base](https://huggingface.co/Salesforce/blip-vqa-base) on the [molvision/Clintox-V-SMILES-0](https://huggingface.co/datasets/molvision/Clintox-V-SMILES-0) dataset for predicting clinical toxicity of molecular compounds.
## Model Description
BLIP2-Clintox-Finetuned is a vision-language model designed to predict the clinical toxicity of chemical compounds. It processes molecular structure images (generated from SMILES representations) to classify compounds based on their toxicity profiles.
## Intended Use
This model is intended for:
- Predicting clinical toxicity of drug compounds
- Assisting in early-stage drug discovery and safety assessment
- Research in computational chemistry and pharmacology
## Training Data
The model was fine-tuned on the Clintox dataset, which contains:
- Molecular structure images generated from SMILES notation
- Binary toxicity labels based on clinical trial outcomes
