LEP-AD

Closed weights King Abdullah University of Science and Technology (KAUST),Karolinska Institute 3B parameters March 2023

No estimate

No hardware requirements for this model

The weights for this model have not been published, so it cannot be downloaded or run on your own hardware at any size. It is reachable only through its provider, and no graphics card changes that.

On record

Full specification

Everything on record for this model. Most of it describes how it was trained rather than how it runs — useful context for judging how much work went into it, and how it compares with models built at a different scale.

Origin

Who built this model, where, and when it was published.

Organisation
King Abdullah University of Science and Technology (KAUST),Karolinska Institute
Organisation type
Academia,Academia
Country
Saudi Arabia, Sweden
Published
15 March 2023
Authors
Anuj Daga, Sumeer Ahmad Khan, David Gomez Cabrero, Robert Hoehndorf, Narsis A. Kiani, Jesper Tegner

What it does

The problem areas the model was built for. A model can carry several of each.

Domain
Biology
Task
Proteins, Protein interaction prediction, Drug discovery, Protein-ligand binding affinity prediction
Base model
ESM2-3B
Numerical format
FP16

Size

How large the model is and how much data it was trained on. Parameters are the figure that decides whether it fits on a given graphics card.

Parameters
3B

Uses ESM-2 3B. Table 2 gives details on the non-ESM layers. The GCN appears to have about 3.31M parameters and the linear layers should have 771k and 3.3M, respectively. So total is ~3.007B

Training data
1,244,420 tokens

Largest dataset appears to be STITCH, at 1244420 drug-target pairs.

Availability

Whether you can obtain the model and run it on your own hardware, which is what decides if any of the graphics-card figures on this page apply.

Weights
Closed — provider access only
Model access
Unreleased
Training code
Open (non-commercial)

https://github.com/adaga06/LEP-AD unclear license

How it is classified

Labels the source dataset applies when tracking notable models, and how confident it is in the entry.

Why it is tracked
SOTA improvement

Tables 3-5 "We report new best-in-class state-of-the-art results compared to competing methods such as SimBoost, DeepCPI, Attention-DTA, GraphDTA, and more using multiple datasets, including Davis, KIBA, DTC, Metz, ToxCast, and STITCH. Finally, we find that a pre-trained model with embedding of proteins (the LED-AD) outperforms a model using an explicit alpha-fold 3D representation of proteins (e.g., LEP-AD supervised by Alphafold)"

Record confidence
Confident
Citations
1

Sources

Where this record came from and when it was last checked.

Reference
LEP-AD: Language Embedding of Proteins and Attention to Drugs predicts Drug Target Interactions
Last updated
28 November 2025

What the numbers mean

What this model is

LEP-AD was published by King Abdullah University of Science and Technology (KAUST),Karolinska Institute, in the country recorded as Saudi Arabia, during March 2023. The publishing organisation is categorised as academia,Academia.

It works in the domain of Biology, and is recorded as performing the task of proteins, Protein interaction prediction, Drug discovery, Protein-ligand binding affinity prediction.

Rather than being trained from scratch, it is derived from ESM2-3B. Most models at this scale are adapted from an existing base rather than built from nothing.

Because the weights are not available, none of the hardware figures elsewhere on this site apply to it.

What went into building it

Training consumed a corpus of around 1,244,420 tokens of text.

The reason it appears in this catalogue at all: sOTA improvement.

Answers

LEP-AD — common questions

01

LEP-AD— how many parameters does it have?

It has a parameter count of 3B. Uses ESM-2 3B. Table 2 gives details on the non-ESM layers. The GCN appears to have about 3.31M parameters and the linear layers should have 771k and 3.3M, respectively. So total is ~3.007B. That figure is the total, and it is what decides how much memory the model needs — roughly half a gigabyte per billion at the compression most people use.

02

LEP-AD— who created it?

It was published by King Abdullah University of Science and Technology (KAUST),Karolinska Institute, based in Saudi Arabia, an organisation categorised as academia,Academia.

03

LEP-AD— when was it released?

It was published in March 2023. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.

04

LEP-AD— what is it used for?

It works in the domain of Biology, and is recorded as handling the task of proteins, Protein interaction prediction, Drug discovery, Protein-ligand binding affinity prediction. A model can carry several of each, so these are the areas it was built for rather than a limit on what it will attempt.

05

LEP-AD— what GPU do I need to run it?

None. This is a closed model — its weights were never published, so it cannot be downloaded or run on your own hardware at any price. It is reachable only through its provider.

06

LEP-AD— is it open source?

No. Its weights have not been published, so it exists only as a service controlled by its owner.

Source

Original publication

Record last updated 28 November 2025

The other direction

Looking at it from the other side?

This page starts from the model. If you already own a card and want to know everything it will run, start from the hardware instead.