ProtT3

Closed weights National University of Singapore,University of Science and Technology of China (USTC),Hokkaido University 1.6B parameters May 2024

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
National University of Singapore,University of Science and Technology of China (USTC),Hokkaido University
Organisation type
Academia,Academia,Academia
Country
Singapore, China, Japan
Published
21 May 2024
Authors
Zhiyuan Liu, An Zhang, Hao Fei, Enzhi Zhang, Xiang Wang, Kenji Kawaguchi, Tat-Seng Chua

What it does

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

Domain
Biology
Task
Protein question answering

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
1.6B

Table 12

Training data
tokens

430,595 + 422,315 + 3,360,000 = 4,212,910 datapoints 4,212,910 × 300 = 1,263,873,000 tokens Final estimate: 1.3 billion tokens Exact numbers from Table 2 430595*(336+48)+422315*(338+101)+3359693*(291+10)=1362012358

Epochs
60

The training run

What it physically took to train: which chips, how many, for how long, and what that drew from the wall.

Training hardware
NVIDIA A100,NVIDIA V100
Chips used
6

How it is classified

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

Record confidence
Likely
Citations
34

Sources

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

Reference
ProtT3: Protein-to-Text Generation for Text-based Protein Understanding
Last updated
25 May 2026

What the numbers mean

What this model is

ProtT3 was published by National University of Singapore,University of Science and Technology of China (USTC),Hokkaido University, in Singapore, in May 2024. academia,Academia,Academia is the category the publisher falls under.

It works in Biology, and is recorded as doing protein question answering.

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

Answers

ProtT3 — common questions

01

When was ProtT3 released?

ProtT3 was published in May 2024. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.

02

What is ProtT3 used for?

ProtT3 works in Biology, and is recorded as handling protein question answering. These are the areas it was designed around; they describe intent rather than a hard boundary.

03

What GPU do I need to run ProtT3?

None. ProtT3 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.

04

Is ProtT3 open source?

The licensing for ProtT3 was never recorded in our source data. We treat unstated licensing as closed, because an unrecorded licence is not one to rely on.

05

How many parameters does ProtT3 have?

ProtT3 has 1.6B parameters. Table 12. 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.

06

Who created ProtT3?

ProtT3 was published by National University of Singapore,University of Science and Technology of China (USTC),Hokkaido University, based in Singapore, categorised as academia,Academia,Academia.

Source

Original publication

Record last updated 25 May 2026

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.