PepPrCLIP
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
- Duke University,Cornell University,Sanford Burnham Prebys Institute
- Organisation type
- Academia,Academia
- Country
- United States of America
- Published
- 22 July 2024
- Authors
- Suhaas Bhat, Kalyan Palepu, Lauren Hong, Joey Mao, Tianzheng Ye, Rema Iyer, Lin Zhao, Tianlai Chen, Sophia Vincoff, Rio Watson, Tian Wang, Divya Srijay, Venkata Srikar Kavirayuni, Kseniia Kholina, Shrey Goel, Pranay Vure, Aniruddha J Desphande, Scott H Soderling, Matthew P DeLisa, Pranam Chatterjee
What it does
The problem areas the model was built for. A model can carry several of each.
- Domain
- Biology
- Task
- Protein design
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
- tokens
- Epochs
- 56
11,597 datapoints from noisy dataset training pairs used for pre-training 11,597 x 1 epoch = 11,597 total datapoints Final estimate: 11,597 ≈ 1.16e4
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
- 1 × 10¹⁸ FLOP
- How it was established
- Hardware
1. Hardware: 1x NVIDIA A100 80GB GPU (3.12e14 FLOP/s fp16) 2. Training duration: 6 hours (estimated based on dataset size and epochs) 3. Utilization: 40% 4. Calculation: 3.12e14 FLOP/s × 0.4 × (6 × 3600s) = 2.69e18 FLOP ≈ 1.0e18 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 A100 SXM4 80 GB
- Chips used
- 1
- Power draw
- 434 W
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 (non-commercial)
- Training code
- Open (non-commercial)
- Hugging Face
- ubiquitx
All code and Colab notebooks to run PepPrCLIP can be freely accessed by the academic community at: https://huggingface.co/ubiquitx/pepprclip, after signing a non-commercial, research-only academic license. All raw and processed data have been deposited to the Zenodo repository: https://zenodo.org/doi/10.5281/zenodo.10971077
How it is classified
Labels the source dataset applies when tracking notable models, and how confident it is in the entry.
- Record confidence
- Speculative
Sources
Where this record came from and when it was last checked.
- Reference
- De Novo Design of Peptide Binders to Conformationally Diverse Targets with Contrastive Language Modeling
- Last updated
- 28 November 2025
What the numbers mean
Background
PepPrCLIP was published by Duke University,Cornell University,Sanford Burnham Prebys Institute, in United States of America, in July 2024. academia,Academia is the category the publisher falls under.
It works in Biology, and is recorded as doing protein design.
Published weights mean the model runs on your machine rather than someone else's, which is what makes the hardware question below answerable at all. It is published under the ubiquitx organisation on Hugging Face.
Training and provenance
Producing it required around 1 × 10¹⁸ FLOP of arithmetic, on NVIDIA A100 SXM4 80 GB, which is a statement about the training budget rather than about inference.
Answers
PepPrCLIP — common questions
When was PepPrCLIP released?
PepPrCLIP was published in July 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.
What is PepPrCLIP used for?
PepPrCLIP works in Biology, and is recorded as handling protein design. These are the areas it was designed around; they describe intent rather than a hard boundary.
Where can I download PepPrCLIP?
Its weights are published under the ubiquitx organisation on Hugging Face. We do not host model files — this site calculates what hardware is needed to run them.
How much compute was used to train PepPrCLIP?
Around 1 × 10¹⁸ FLOP, on NVIDIA A100 SXM4 80 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.
What GPU do I need to run PepPrCLIP?
We cannot say. PepPrCLIP 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 PepPrCLIP open source?
Its weights are published, so PepPrCLIP 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 PepPrCLIP have?
No parameter count has been published for PepPrCLIP, which is why no memory or speed figure appears on this page.
Who created PepPrCLIP?
PepPrCLIP was published by Duke University,Cornell University,Sanford Burnham Prebys Institute, based in United States of America, categorised as academia,Academia.
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.