ESM3 (98B)
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
- EvolutionaryScale,University of California (UC) Berkeley
- Organisation type
- Industry,Academia
- Country
- United States of America
- Published
- 25 June 2024
- Authors
- Thomas Hayes, Roshan Rao, Halil Akin, Nicholas James Sofroniew, Deniz Oktay, Zeming Lin, Robert Verkuil, Vincent Quy Tran, Jonathan Deaton, Marius Wiggert, Rohil Badkundri, Irhum Shafkat, Jun Gong, Alexander Derry, Raul Santiago Molina, Neil Thomas, Yousuf Khan, Chetan Mishra, Carolyn Kim, Liam J Bartie, Patrick D Hsu, Tom Sercu, Salvatore Candido, Alexander Rives
What it does
The problem areas the model was built for. A model can carry several of each.
- Domain
- Biology
- Task
- Protein generation
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
- 98.5B
- Training data
- 771,000,000,000 tokens
- Epochs
- 2.3
- Batch size
- 4,194,304
98.5 billion (Table S1)
771 billion tokens
Table S1
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
- 1.1 × 10²⁴ FLOP
- How it was established
- Reported
"ESM3 at its largest scale was trained with 1.07×10^24 FLOPs on 2.78 billion proteins and 771 billion unique tokens, and has 98 billion parameters." per Table 1, trained 98B model on 1.8T training tokens. 98 billion * 1800 billion * 6 = 1.06e24. Likely some rounding, so will go with developer's reported count.
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
only small version released
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
- Why it is tracked
- Historical significance
- Record confidence
- Confident
Largest (in compute) biology and protein model to date, was able to discover novel green fluorescent proteins
Sources
Where this record came from and when it was last checked.
- Reference
- ESM3: Simulating 500 million years of evolution with a language model
- Last updated
- 11 February 2026
What the numbers mean
What this model is
ESM3 (98B) was published by EvolutionaryScale,University of California (UC) Berkeley, in the country recorded as United States of America, during June 2024. It comes out of an organisation categorised as industry,Academia.
It works in the domain of Biology, and is recorded as performing the task of protein generation.
Because the weights are not available, none of the hardware figures elsewhere on this site apply to it.
Training and provenance
Producing it required arithmetic totalling around 1.1 × 10²⁴ FLOP. That figure measures what producing the model cost, and has no bearing on how fast it answers.
It was trained on a corpus of about 771,000,000,000 tokens of text.
The reason it appears in this catalogue at all: historical significance.
Answers
ESM3 (98B) — common questions
ESM3 (98B)— who created it?
It was published by EvolutionaryScale,University of California (UC) Berkeley, based in United States of America, an organisation categorised as industry,Academia.
ESM3 (98B)— when was it released?
It was published in June 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.
ESM3 (98B)— what is it used for?
It works in the domain of Biology, and is recorded as handling the task of protein generation. These are the areas it was designed around; they describe intent rather than a hard boundary.
ESM3 (98B)— how much compute was used to train it?
Training consumed around 1.1 × 10²⁴ 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.
ESM3 (98B)— what GPU do I need to run it?
None. This 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.
ESM3 (98B)— is it open source?
No. Its weights have not been published, so it exists only as a service controlled by its owner.
ESM3 (98B)— how many parameters does it have?
It has a parameter count of 98.5B. 98.5 billion (Table S1). 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.
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.