AIbase
EN
Home
Esm2 T48 15B UR50D
E

Esm2 T48 15B UR50D

Developed by facebook
ESM-2 is a state-of-the-art protein model trained with masked language modeling objectives, suitable for fine-tuning on various protein sequence tasks.
Protein Model
Transformers
Open Source License:MIT#Protein Sequence Prediction#Masked Language Modeling#Biomedical Research
Downloads 20.80k
Release Time : 9/26/2022

Model Overview

ESM-2 is a protein language model trained via masked language modeling objectives, capable of understanding and generating protein sequences, and applicable to multiple bioinformatics tasks.

Model Features

Large-scale Pretraining
The model is pretrained on extensive protein sequence data, exhibiting robust sequence understanding and generation capabilities.
Multi-scale Models
Offers models ranging from 8 million to 15 billion parameters to accommodate varying computational resources and precision needs.
Easy Fine-tuning
Designed for fine-tuning on specific protein tasks, supporting diverse downstream applications.

Model Capabilities

Protein sequence generation
Protein structure prediction
Protein function annotation
Protein sequence classification

Use Cases

Bioinformatics
Protein Function Prediction
Predicts potential functions by analyzing protein sequences.
High-precision function annotation
Protein Structure Prediction
Predicts the 3D structure of proteins from amino acid sequences.
Structure predictions approaching experimental accuracy
Drug Discovery
Protein Design
Generates protein sequences with specific functions.
Novel protein designs
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
EnglishįŽ€äŊ“中文įšéĢ”中文ãĢãģんご
Š 2025AIbase