---
license: llama2
datasets:
- mims-harvard/ProCyon-Instruct
language:
- en
base_model:
- meta-llama/Llama-2-7b-hf
tags:
- biology
- protein
---
# ProCyon-Split

ProCyon-Split is a multimodal foundation model for protein phenotypes, which combines a large language model with protein encoders to support inputs of interleaved free text and proteins.
In contrast to ProCyon-Full, this model is instruction-tuned using the training split of the [ProCyon-Instruct](https://huggingface.co/datasets/mims-harvard/ProCyon-Instruct) dataset to
enable rigorous model evaluation on held-out protein-phenotype pairs.

For more information on the model design, training, and validation, please see the [overview page](https://zitniklab.hms.harvard.edu/ProCyon/) or the [paper](https://www.biorxiv.org/content/10.1101/2024.12.10.627665v1).

Additional versions of the model are available as [ProCyon-Full](https://huggingface.co/mims-harvard/ProCyon-Full) and [ProCyon-Bind](https://huggingface.co/mims-harvard/ProCyon-Bind).