SaProt

Closed weights Zhejiang University (ZJU),Westlake University 650M parameters April 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
Zhejiang University (ZJU),Westlake University
Organisation type
Academia,Academia
Country
China
Published
19 April 2024
Authors
Jin Su, Chenchen Han, Yuyang Zhou, Junjie Shan, Xibin Zhou, Fajie Yuan

What it does

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

Domain
Biology
Task
Protein or nucleotide language model (pLM/nLM)

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
650M

"To our knowledge, SaProt stands out as the PLM currently trained with the largest number of protein structures, containing 650 million parameters" Assume 40% utilization, FP16 tensor precision

Training data
tokens

4.0 × 10⁷ sequences × 3.0 × 10² tokens/sequence = 1.2 × 10¹⁰ tokens

Training compute

The arithmetic performed to train the model, measured in floating-point operations. It is a measure of what the training run cost, not of how fast the finished model answers you.

Training compute
6.2 × 10²² FLOP

"Its training lasted 3 months and utilized 64 NVIDIA 80G A100 GPUs," 64*312000000000000*0.4*3months=6.210847e+22

How it was established
Hardware

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
Chips used
64
Wall-clock time
2,160 hours (90 days)

"Its training lasted 3 months "

Power draw
50.6 kW

How it is classified

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

Record confidence
Confident
Citations
229

Sources

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

Reference
SaProt: Protein Language Modeling with Structure-aware Vocabulary
Last updated
1 January 2026

What the numbers mean

About this model

SaProt was published by Zhejiang University (ZJU),Westlake University, in China, in April 2024. academia,Academia is the category the publisher falls under.

It works in Biology, and is recorded as doing protein or nucleotide language model (pLM/nLM).

Its weights were never published, so it can only be reached through its provider. No graphics card changes that.

Training and provenance

Training it took roughly 6.2 × 10²² FLOP of computation, on NVIDIA A100 — a measure of what producing the model cost, not of how fast it answers.

Answers

SaProt — common questions

01

How much compute was used to train SaProt?

Around 6.2 × 10²² FLOP, on NVIDIA A100. That measures what producing the model cost and says nothing about how quickly it answers once trained — inference speed comes from memory bandwidth, not from the training budget.

02

What GPU do I need to run SaProt?

None. SaProt 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.

03

Is SaProt open source?

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

04

How many parameters does SaProt have?

SaProt has 650M parameters. "To our knowledge, SaProt stands out as the PLM currently trained with the largest number of protein structures, containing 650 million parameters" Assume 40% utilization, FP16 tensor precision. 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.

05

Who created SaProt?

SaProt was published by Zhejiang University (ZJU),Westlake University, based in China, categorised as academia,Academia.

06

When was SaProt released?

SaProt was published in April 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.

07

What is SaProt used for?

SaProt works in Biology, and is recorded as handling protein or nucleotide language model (pLM/nLM). These are the areas it was designed around; they describe intent rather than a hard boundary.

Source

Original publication

Record last updated 1 January 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.