PeptideBERT
No estimate
No hardware requirements for this model
This model's weights are open, but no parameter count has been published for it. Every memory and speed figure starts from that number, so we would rather show nothing than a fabricated estimate.
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
- Carnegie Mellon University (CMU)
- Organisation type
- Academia
- Country
- United States of America
- Published
- 28 August 2023
- Authors
- Chakradhar Guntuboina, Adrita Das, Parisa Mollaei, Seongwon Kim, and Amir Barati Farimani
What it does
The problem areas the model was built for. A model can carry several of each.
- Domain
- Biology
- Task
- Proteins, Protein property prediction
- Base model
- ProtBERT-UniRef
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.
- Training data
- 4,160,566 tokens
- Epochs
- 30
Pretraining: 217,000,000 sequences × 100 residues = 2.17 × 10¹⁰ tokens Fine-tuning sequences: 9,316 + 29,892 + 17,185 = 56,393 sequences 56,393 × 100 residues = 5.64 × 10⁶ tokens Total tokens: 2.17 × 10¹⁰ + 5.64 × 10⁶ ≈ 2.17 × 10¹⁰
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
- 4.9 × 10¹⁶ FLOP
- How it was established
- Hardware
- Fine-tuning compute
- 5 × 10¹⁶ FLOP
"Compute for fine-tuning ProtBERT: 1 NVidia GeForce GTX 1080Ti, 30 epochs, batch size 32, model trained for individual tasks with training time ranging from 58-116 minutes, assuming from Table 1 we have 244 minutes 11.34e12 FLOPs and 0.3 utilization rate FLOP = 244 min * 60 sec/min * 11.34e12 FLOP/sec *0.3 = 4.9e16 FLOP,
"Compute for fine-tuning ProtBERT: 1 NVidia GeForce GTX 1080Ti, 30 epochs, batch size 32, model trained for individual tasks with training time ranging from 58-116 minutes, assuming from Table 1 we have 244 minutes 11.34e12 FLOPs and 0.3 utilization rate FLOP = 244 min * 60 sec/min * 11.34e12 FLOP/sec *0.3 = 4.9e16 FLOP,
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 GeForce GTX 1080 Ti
- Chips used
- 1
- Chip-hours
- 4
- Wall-clock time
- 4 hours
- Power draw
- 273 W
244 minues from Table 1
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
- Open — downloadable
- Model access
- Open weights (unrestricted)
- Training code
- Open source
MIT (models, training, inference): https://github.com/ChakradharG/PeptideBERT
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
- Record confidence
- Confident
"Our model has achieved state of the art (SOTA) for predicting Hemolysis, which is a task for determining peptide’s potential to induce red blood cell lysis."
Sources
Where this record came from and when it was last checked.
- Reference
- PeptideBERT: A language Model based on Transformers for Peptide Property Prediction
- Last updated
- 28 November 2025
What the numbers mean
Where it came from
PeptideBERT was published by Carnegie Mellon University (CMU), in United States of America, in August 2023. academia is the category the publisher falls under.
It works in Biology, and is recorded as doing proteins, Protein property prediction.
It builds on ProtBERT-UniRef, which is why it shares that model's general shape and size.
The weights being open is what puts this page in the calculator rather than only in the catalogue: it is a model you can actually hold.
Training and provenance
Training it took roughly 4.9 × 10¹⁶ FLOP of computation, on NVIDIA GeForce GTX 1080 Ti — a measure of what producing the model cost, not of how fast it answers.
It was trained on about 4,160,566 tokens of text.
Its inclusion criterion is sOTA improvement.
Answers
PeptideBERT — common questions
How much compute was used to train PeptideBERT?
Around 4.9 × 10¹⁶ FLOP, on NVIDIA GeForce GTX 1080 Ti. 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 PeptideBERT?
We cannot say. PeptideBERT has open weights, but no parameter count has been published for it, and every memory and speed calculation starts from that number. We would rather show nothing than a fabricated estimate.
Is PeptideBERT open source?
Its weights are published, so PeptideBERT can be downloaded and run on your own hardware. Note that open weights is not the same as open source in the full sense — it says nothing about the training data, the training code, or the commercial terms attached.
How many parameters does PeptideBERT have?
No parameter count has been published for PeptideBERT, which is why no memory or speed figure appears on this page.
Who created PeptideBERT?
PeptideBERT was published by Carnegie Mellon University (CMU), based in United States of America, categorised as academia.
When was PeptideBERT released?
PeptideBERT was published in August 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.
What is PeptideBERT used for?
PeptideBERT works in Biology, and is recorded as handling proteins, Protein property 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.
Where can I download PeptideBERT?
The weights for PeptideBERT are published, though we do not hold a repository link for it. This site calculates hardware requirements rather than hosting model files.
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.