The Best 59 Protein Model Tools in 2025
Esm2 T36 3B UR50D
MIT
ESM-2 is a next-generation protein model trained with masked language modeling objectives, suitable for fine-tuning on various downstream tasks with protein sequences as input.
Protein Model
Transformers

E
facebook
3.5M
22
Esm2 T6 8M UR50D
MIT
ESM-2 is a next-generation protein model trained with masked language modeling objectives, suitable for fine-tuning on various protein sequence tasks.
Protein Model
Transformers

E
facebook
1.5M
21
Esm2 T33 650M UR50D
MIT
ESM-2 is a state-of-the-art protein model trained on masked language modeling objectives, suitable for protein sequence analysis and prediction tasks
Protein Model
Transformers

E
facebook
640.23k
41
Esm2 T12 35M UR50D
MIT
ESM-2 is a cutting-edge protein model trained on masked language modeling objectives, suitable for various protein sequence analysis tasks
Protein Model
Transformers

E
facebook
332.83k
15
Prot Bert
A protein sequence pre-training model based on BERT architecture, capturing biophysical properties of protein sequences through self-supervised learning
Protein Model
Transformers

P
Rostlab
276.10k
111
Prostt5
MIT
ProstT5 is a protein language model capable of translating between protein sequences and structures.
Protein Model
Transformers

P
Rostlab
252.91k
23
Prot T5 Xl Uniref50
A protein sequence pre-training model based on T5-3B architecture that captures biophysical properties through self-supervised learning
Protein Model
Transformers

P
Rostlab
78.45k
44
Esm2 T30 150M UR50D
MIT
ESM-2 is a state-of-the-art protein model trained on masked language modeling objectives, suitable for fine-tuning on various protein sequence input tasks.
Protein Model
Transformers

E
facebook
69.91k
7
Prot Bert Bfd
A BERT-based protein sequence pre-training model that extracts biophysical features through self-supervised learning from 2.1 billion protein sequences
Protein Model
Transformers

P
Rostlab
30.60k
16
Esm1b T33 650M UR50S
MIT
ESM-1b is a Transformer-based protein language model trained via unsupervised learning on protein sequence data, capable of predicting protein structure and function.
Protein Model
Transformers

E
facebook
24.20k
18
Rinalmo
RiNALMo is a non-coding RNA (ncRNA) model pre-trained based on the masked language modeling (MLM) objective, trained through self-supervised learning on a large number of non-coding RNA sequences.
Protein Model Other
R
multimolecule
21.38k
2
Rna Torsionbert
Other
A BERT-based model for predicting RNA torsion angles and pseudo-torsion angles from RNA sequences
Protein Model
Transformers

R
sayby
20.86k
7
Esm2 T48 15B UR50D
MIT
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

E
facebook
20.80k
20
Protgpt2
Apache-2.0
ProtGPT2 is a protein language model based on the GPT2 architecture, capable of generating novel protein sequences while retaining key features of natural proteins.
Protein Model
Transformers

P
nferruz
17.99k
108
Helix Mrna
Helix-mRNA is a hybrid model combining state-space and transformer architectures, specifically designed for mRNA sequence analysis
Protein Model
Transformers Other

H
helical-ai
10.32k
10
Esmplusplus Large
ESM++ is a faithful implementation of ESMC, supporting batch operations and fully compatible with the Huggingface ecosystem without relying on the official ESM Python package. The large version corresponds to ESMC's 600-million parameter model.
Protein Model
Transformers

E
Synthyra
8,314
9
DRP
MIT
DRP is a model focused on robot motion planning, trained for 7k epochs (approximately 7M steps) under a 5M checkpoint
Protein Model English
D
jimyoung6709
7,980
1
Rnafm
RNA foundation model pre-trained on non-coding RNA data using masked language modeling (MLM) objective
Protein Model
Safetensors Other
R
multimolecule
6,791
1
Esmplusplus Small
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.
Protein Model
Transformers

E
Synthyra
6,460
14
Saprot 650M AF2
MIT
SaProt is a pre-trained model based on protein sequence and structural information, specifically optimized for low pLDDT regions.
Protein Model
Transformers

S
westlake-repl
5,630
14
Protein Matryoshka Embeddings
CC
This model generates embedding vectors for protein sequences, supporting shortened embeddings to accelerate search tasks.
Protein Model
Transformers

P
monsoon-nlp
2,121
7
Distilprotbert
MIT
Distilled version of ProtBert-UniRef100 model for protein feature extraction and downstream task fine-tuning
Protein Model
Transformers

D
yarongef
1,965
9
Bert Protein Classifier
This model is fine-tuned based on Bert-Base-Uncased and is used for the multi-label classification task of predicting protein functions based on protein amino acid sequences.
Protein Model
Transformers

B
oohtmeel
1,772
1
Generator Eukaryote 3b Base
MIT
GENERator is a generative genome foundation model with a 98,000 base pair context length and 3 billion parameters, trained on an extended dataset of eukaryotic DNA
Protein Model
Transformers

G
GenerTeam
1,599
5
Codontransformer
Apache-2.0
The ultimate tool for codon optimization, capable of converting protein sequences into DNA sequences optimized for target organisms.
Protein Model
Transformers

C
adibvafa
1,327
7
Prot T5 Base Mt Uniref50
This model is used to predict mass spectrometry data for given amino acid sequences.
Protein Model
P
Rostlab
671
0
Prollama Stage 1
Apache-2.0
ProLLaMA is a protein large language model based on the Llama-2-7b architecture, specializing in multitask protein language processing.
Protein Model
Transformers

P
GreatCaptainNemo
650
2
Prot T5 Xl Bfd
ProtT5-XL-BFD is a self-supervised pre-trained model based on protein sequences, using the T5 architecture, trained on 2.1 billion protein sequences for protein feature extraction and downstream task fine-tuning.
Protein Model
Transformers

P
Rostlab
605
10
Gobert
MIT
GoBERT is a model specifically designed for general gene function prediction. It can effectively capture the relationships between Gene Ontology (GO) functions by leveraging gene ontology graph information.
Protein Model
Safetensors
G
MM-YY-WW
479
1
Gpn Msa Sapiens
MIT
GPN-MSA is a language model trained on human and 89 other vertebrate species, specializing in DNA sequence analysis and variant effect prediction.
Protein Model
Transformers Other

G
songlab
446
8
Pepmlm 650M
The first de novo generator of linear peptide binders dependent solely on target protein sequences
Protein Model
Transformers

P
ChatterjeeLab
396
19
Prollama
Apache-2.0
ProLLaMA is a protein large language model based on Llama-2-7b, specializing in multitask protein language processing, capable of performing protein sequence generation and superfamily classification tasks.
Protein Model
Transformers

P
GreatCaptainNemo
366
16
HIV V3 Bodysite
Optimized model based on HIV-BERT for predicting the collection site of HIV V3 loop samples, with significantly improved accuracy compared to the original model
Protein Model
Transformers

H
damlab
257
0
ESM2 35M Protein Molecular Function
A protein function prediction tool based on the Evolutionary Scale Modeling (ESM), which predicts the molecular function of proteins through amino acid sequences.
Protein Model
Transformers

E
andrewdalpino
254
1
ESM2 150M Protein Molecular Function
A protein molecular function prediction model based on the Gene Ontology (GO) and the ESM2 architecture
Protein Model
Transformers

E
andrewdalpino
175
1
Protst Esm1b
The ProtST framework enhances the pre-training and understanding of protein sequences through biomedical text, constructs the ProtDescribe dataset, designs three pre-training tasks, and supports supervised learning and zero-shot prediction.
Protein Model
Transformers

P
mila-intel
173
1
Isoformer
Isoformer is a model capable of accurately predicting differential transcript expression, outperforming existing methods and fully leveraging multimodal data.
Protein Model
Transformers

I
InstaDeepAI
165
3
Protgpt2 Distilled Tiny
Apache-2.0
A distilled version of ProtGPT2, compressed into a more efficient small model through knowledge distillation, maintaining performance while improving inference speed
Protein Model
Transformers

P
littleworth
157
4
Ablang Heavy
AbLang is an antibody language model specifically designed for processing antibody sequences, particularly heavy chain sequences. The model is trained on uppercase amino acid letters and is suitable for protein feature extraction and downstream task fine-tuning.
Protein Model
Transformers

A
qilowoq
131
6
Ankh3 Xl
Ankh3 is a protein language model based on the T5 architecture. It is pre - trained by jointly optimizing masked language modeling and sequence completion tasks, and is suitable for protein feature extraction and sequence analysis.
Protein Model
Transformers

A
ElnaggarLab
131
2
Nanobert
nanoBERT is a nanobody-specific transformer for predicting amino acids at specific positions in a given query sequence.
Protein Model
Transformers

N
NaturalAntibody
117
4
Segment Nt Multi Species
SegmentNT-multi-species is a segmentation model based on the Nucleotide Transformer, designed to predict the locations of various genomic elements at single-nucleotide resolution.
Protein Model
Transformers

S
InstaDeepAI
102
1
- 1
- 2