| # DPLM | |
| DPLM (diffusion protein language model) is a versatile protein language model that demonstrates strong generative and predictive capabilities for protein sequences. Specifically, DPLM exhibits impressive performance in protein sequence generation, motif scaffolding, inverse folding, and representation learning. | |
| For more detailed information about DPLM, please refer to our paper [Diffusion Language Models Are Versatile Protein Learners](https://arxiv.org/abs/2402.18567). | |
| This repository contains the DPLM model checkpoint of 3B parameters. | |
| Please refer to our [github repository](https://github.com/bytedance/dplm/tree/main) for code and usage. | |
| For example, you can load DPLM model as below: | |
| ``` | |
| from byprot.models.lm.dplm import DiffusionProteinLanguageModel | |
| model_name = "airkingbd/dplm_3B" | |
| dplm = DiffusionProteinLanguageModel.from_pretrained(model_name) | |
| ``` | |
| All DPLM checkpoints are available in the table below: | |
| | Model size | Num layers | Num parameters | | |
| |------------------------------|----|----------| | |
| | [dplm_3b](https://huggingface.co/airkingbd/dplm_3b) | 36 | 3B | | |
| | [dplm_650m](https://huggingface.co/airkingbd/dplm_650m) | 33 | 650M | | |
| | [dplm_150m](https://huggingface.co/airkingbd/dplm_150m) | 30 | 150M | | |
| **News**: welcome to check our new work [DPLM-2: A Multimodal Diffusion Protein Language Model](https://huggingface.co/papers/2410.13782), a multimodal protein foundation model that extends DPLM to simultaneously model, understand, and generate both sequences and structures! |