E

Esmplusplus Small

Developed by Synthyra
ESM++ is a faithful implementation of ESMC, supporting batch processing and compatible with the standard Huggingface interface without requiring the ESM Python package. The small version corresponds to the 300-million-parameter version of ESMC.
Downloads 6,460
Release Time : 12/4/2024

Model Overview

ESM++ is a protein language model designed for masked language modeling, sequence classification, and token classification tasks on protein sequences.

Model Features

Efficient batch processing
Compared to ESMC, ESM++ significantly improves throughput with efficient batch processing, even faster at a batch size of 1.
Compatible with Huggingface interface
Fully compatible with the standard Huggingface interface, no dependency on the ESM Python package required.
Multi-precision support
Supports fp32, fp16, and bf16 precision; the fp16 version is closer to fp32 output and is recommended for use.
Fast embedding
Provides the embed_dataset method for quickly embedding entire protein sequence datasets.

Model Capabilities

Protein sequence embedding
Masked language modeling
Sequence classification
Token classification
Attention map generation

Use Cases

Protein research
Protein function prediction
Use sequence classification to predict protein function.
Protein structure prediction
Utilize protein sequence embeddings for structure prediction.
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
Š 2025AIbase