DiffPepBuilder

Closed weights Peking University 104M 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
Peking University
Organisation type
Academia
Country
China
Published
30 April 2024
Authors
Fanhao Wang, Yuzhe Wang, Laiyi Feng, Changsheng Zhang, Luhua Lai

What it does

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

Domain
Biology
Task
Protein generation
Base model
ESM2-650M

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
104M
Training data
1,489,700 tokens

Step-by-step calculation: Complexes = 14,897 Average length = (8 + 30)/2 = 19 Total datapoints = 14,897 × 19 = 283,043 ≈ 2.83×10⁵

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
7.7 × 10²¹ FLOP

1. Hardware: 8x NVIDIA A800 80GB GPUs (7.80e13 FLOP/s per GPU) 2. Training duration: 5 days = 432,000 seconds (directly provided) 3. Utilization: 40% 4. Calculation: 7.80e13 FLOP/s/GPU × 8 GPUs × 432,000s × 0.40 = 1.08e20 FLOP (7.80e13 × 8 = 6.24e14 FLOP/s total → × 432,000s = 2.70e20 → × 0.40 = 1.08e20) Base model: 7.560000000001e+21 Total: 7667785728001000000000

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 A800 PCIe 40 GB
Chips used
8
Wall-clock time
120 hours
Power draw
4.0 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
27

Sources

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

Reference
Target-Specific De Novo Peptide Binder Design with DiffPepBuilder
Last updated
25 May 2026

What the numbers mean

Background

DiffPepBuilder was published by Peking University, in China, in April 2024. academia is the category the publisher falls under.

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

It builds on ESM2-650M, which is why it shares that model's general shape and size.

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

What went into building it

The training run consumed about 7.7 × 10²¹ FLOP, on NVIDIA A800 PCIe 40 GB. That figure describes the cost of creating it and has no bearing on how quickly it generates text.

It was trained on about 1,489,700 tokens of text.

Answers

DiffPepBuilder — common questions

01

Who created DiffPepBuilder?

DiffPepBuilder was published by Peking University, based in China, categorised as academia.

02

When was DiffPepBuilder released?

DiffPepBuilder 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.

03

What is DiffPepBuilder used for?

DiffPepBuilder works in Biology, and is recorded as handling protein generation. 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.

04

How much compute was used to train DiffPepBuilder?

Around 7.7 × 10²¹ FLOP, on NVIDIA A800 PCIe 40 GB. 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.

05

What GPU do I need to run DiffPepBuilder?

None. DiffPepBuilder 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

Is DiffPepBuilder open source?

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

07

How many parameters does DiffPepBuilder have?

DiffPepBuilder has 104M parameters. 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.

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.