ProtChatGPT
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
- University of Technology Sydney,Zhejiang University (ZJU)
- Organisation type
- Academia,Academia
- Country
- Australia, China
- Published
- 15 February 2024
- Authors
- Chao Wang, Hehe Fan, Ruijie Quan, Yi Yang
What it does
The problem areas the model was built for. A model can carry several of each.
- Domain
- Biology
- Task
- Protein question answering
- Base model
- Llama 3-8B
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
- 8B
- Training data
- tokens
Stage 1: - Proteins: 549,000 × 500 = 2.745 × 10⁸ - Descriptions: 549,000 × 50 = 2.745 × 10⁷ - Total: 2.745 × 10⁸ + 2.745 × 10⁷ = 3.0195 × 10⁸ Stage 2: - Proteins: 143,508 × 500 = 7.1754 × 10⁷ - Descriptions: 143,508 × 50 = 7.1754 × 10⁶ - Total: 7.1754 × 10⁷ + 7.1754 × 10⁶ = 7.892 × 10⁷ Final Total: 3.0195 × 10⁸ + 7.892 × 10⁷ = 3.8087 × 10⁸
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
- 7.2 × 10²³ FLOP
- How it was established
- Hardware
- Fine-tuning compute
- 3.5 × 10¹⁹ FLOP
GPU hours: 86400s*9.9e+14*2*0.4=6.8e+19 6ND: 6*8000000000*380000001=18240000048000000000 GMean(1.8e+19,6.8e+19)=3.5e+19 Base model: 7.2e+23 Total: 720035320120906900000000
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 H100 SXM5 80GB
- Chips used
- 2
- Wall-clock time
- 24 hours
- Power draw
- 2.8 kW
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
- Unreleased
"Code and our pre-trained model will be publicly available" as of May'25 I cannot find it
How it is classified
Labels the source dataset applies when tracking notable models, and how confident it is in the entry.
- Likely above 10²³ FLOP
- Yes
- Record confidence
- Confident
- Citations
- 23
Sources
Where this record came from and when it was last checked.
- Reference
- ProtChatGPT: Towards Understanding Proteins with Large Language Models
- Last updated
- 25 May 2026
What the numbers mean
Background
ProtChatGPT was published by University of Technology Sydney,Zhejiang University (ZJU), in Australia, in February 2024. It comes out of academia,Academia.
It works in Biology, and is recorded as doing protein question answering.
It is derived from Llama 3-8B rather than trained from scratch, which is the usual way a specialised model is produced.
This is a closed model: the trained values stayed with whoever produced them, and there is no local version to run.
What went into building it
Producing it required around 7.2 × 10²³ FLOP of arithmetic, on NVIDIA H100 SXM5 80GB, which is a statement about the training budget rather than about inference.
Answers
ProtChatGPT — common questions
What is ProtChatGPT used for?
ProtChatGPT works in Biology, and is recorded as handling protein question answering. Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.
How much compute was used to train ProtChatGPT?
Around 7.2 × 10²³ FLOP, on NVIDIA H100 SXM5 80GB. 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.
What GPU do I need to run ProtChatGPT?
None. ProtChatGPT 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.
Is ProtChatGPT open source?
No. ProtChatGPT has not had its weights published, so it exists only as a service controlled by its owner.
How many parameters does ProtChatGPT have?
ProtChatGPT has 8B 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.
Who created ProtChatGPT?
ProtChatGPT was published by University of Technology Sydney,Zhejiang University (ZJU), based in Australia, categorised as academia,Academia.
When was ProtChatGPT released?
ProtChatGPT was published in February 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.
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.