ProGen
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
- Salesforce Research,Stanford University
- Organisation type
- Industry,Academia
- Country
- United States of America
- Published
- 13 March 2020
- Authors
- Ali Madani, Bryan McCann, Nikhil Naik, Nitish Shirish Keskar, Namrata Anand, Raphael R. Eguchi, View ORCID ProfilePo-Ssu Huang, Richard Socher
What it does
The problem areas the model was built for. A model can carry several of each.
- Domain
- Biology
- Task
- Protein generation, Proteins, Protein or nucleotide language model (pLM/nLM)
- Approach
- Self-supervised learning
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
- 1.2B
- Training data
- tokens
- Epochs
- 5
"We train a 1.2B-parameter language model, ProGen, on ∼280M protein sequences"
1,049B from Table 9 https://www.biorxiv.org/content/10.1101/2023.07.05.547496v1
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
- 3.7 × 10²⁰ FLOP
- How it was established
- Hardware,Third-party estimation
Our model was implemented in TensorFlow (Abadi et al., 2016) and trained with a global batch size of 64 distributed across 256 cores of a Cloud TPU v3 Pod for 1M iterations. Training took approximately two weeks using Adagrad (Duchi et al., 2011) 4.00E+12*256*60**2*24*14*0.3 = 3.7e20 7.6E+21 FLOPs from Table 9 https://www.biorxiv.org/content/10.1101/2023.07.05.547496v1
How it is classified
Labels the source dataset applies when tracking notable models, and how confident it is in the entry.
- Record confidence
- Likely
- Citations
- 315
Sources
Where this record came from and when it was last checked.
- Reference
- ProGen: Language Modeling for Protein Generation
- Last updated
- 1 January 2026
What the numbers mean
What this model is
ProGen was published by Salesforce Research,Stanford University, in United States of America, in March 2020. The organisation is categorised as industry,Academia.
It works in Biology, and is recorded as doing protein generation, Proteins, Protein or nucleotide language model (pLM/nLM).
This is a closed model: the trained values stayed with whoever produced them, and there is no local version to run.
Training and provenance
Training it took roughly 3.7 × 10²⁰ FLOP of computation — a measure of what producing the model cost, not of how fast it answers.
Answers
ProGen — common questions
When was ProGen released?
ProGen was published in March 2020. 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 ProGen used for?
ProGen works in Biology, and is recorded as handling protein generation, Proteins, Protein or nucleotide language model (pLM/nLM). 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.
How much compute was used to train ProGen?
Around 3.7 × 10²⁰ FLOP. 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 ProGen?
None. ProGen 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 ProGen open source?
The licensing for ProGen was never recorded in our source data. We treat unstated licensing as closed, because an unrecorded licence is not one to rely on.
How many parameters does ProGen have?
ProGen has 1.2B parameters. "We train a 1.2B-parameter language model, ProGen, on ∼280M protein sequences". 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 ProGen?
ProGen was published by Salesforce Research,Stanford University, based in United States of America, categorised as industry,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.