CaLM
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 Oxford
- Organisation type
- Academia
- Country
- United Kingdom of Great Britain and Northern Ireland
- Published
- 19 December 2022
- Authors
- Carlos Outeiral, Charlotte M. Deane
What it does
The problem areas the model was built for. A model can carry several of each.
- Domain
- Biology
- Task
- Protein or nucleotide language model (pLM/nLM), Protein embedding, Protein property prediction, Protein localization prediction
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
- 86M
- Training data
- 2,523,746,560 tokens
- Epochs
- 14
"We trained a large language model with 86M parameters"
"a dataset of 9M non-redundant and diverse cDNA sequences identified from whole-genome sequencing" "Gradients were accumulated to an effective batch size of 1,000 examples, or approximately 256,000 tokens. " 9000000*256000/1000=2304000000 tokens
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
- 2.9 × 10¹⁹ FLOP
- How it was established
- Hardware,Operation counting
"4 NVIDIA Quadro RTX4000 GPUs for 40 days" Calculation assuming FP32, utilization 30%: = (40 * 24 * 3600) s * 7.1e12 FLOP/s * 0.3 * 4 GPU = 2.999808e+19 alternative calculation: "Gradients were accumulated to an effective batch size of 1,000 examples, or approximately 256,000 tokens. " "(66,000 gradient steps, 14 full epochs)" 256000*66000*14*86000000*6=1.220567e+20
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 Quadro RTX 4000
- Chips used
- 4
- Chip-hours
- 3,840
- Wall-clock time
- 960 hours (40 days)
- Power draw
- 1.3 kW
"The model reported in this work was trained on 4 NVIDIA Quadro RTX4000 GPUs for 40 days (66,000 gradient steps, 14 full epochs)"
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
- Open source
BSD-3-Clause license https://github.com/oxpig/CaLM
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
- Likely
- Citations
- 34
not absolute SOTA, SOTA among similar size models "We show that large language models trained on codons, instead of amino acid sequences, provide high-quality representations that outperform comparable state-of-the-art models across a variety of tasks. In some tasks, like species recognition, prediction of protein and transcript abundance, or melting point estimation, we show that a language model trained on codons outperforms every other published protein language model, including some that co…
Sources
Where this record came from and when it was last checked.
- Reference
- Codon language embeddings provide strong signals for protein engineering
- Last updated
- 1 January 2026
What the numbers mean
What this model is
CaLM was published by University of Oxford, in the country recorded as United Kingdom of Great Britain and Northern Ireland, during December 2022. It comes out of an organisation categorised as academia.
It works in the domain of Biology, and is recorded as performing the task of protein or nucleotide language model (pLM/nLM), Protein embedding, Protein property prediction, Protein localization prediction.
Because the weights are not available, none of the hardware figures elsewhere on this site apply to it.
How it was trained
The training run consumed about 2.9 × 10¹⁹ FLOP, on hardware recorded as NVIDIA Quadro RTX 4000. 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 2,523,746,560 tokens of text.
Its inclusion criterion: sOTA improvement.
Answers
CaLM — common questions
CaLM— who created it?
It was published by University of Oxford, based in United Kingdom of Great Britain and Northern Ireland, an organisation categorised as academia.
CaLM— when was it released?
It was published in December 2022. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.
CaLM— what is it used for?
It works in the domain of Biology, and is recorded as handling the task of protein or nucleotide language model (pLM/nLM), Protein embedding, Protein property prediction, Protein localization prediction. Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.
CaLM— how much compute was used to train it?
Training consumed around 2.9 × 10¹⁹ FLOP, on hardware recorded as NVIDIA Quadro RTX 4000. 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.
CaLM— 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.
CaLM— is it open source?
No. Its weights have not been published, so it exists only as a service controlled by its owner.
CaLM— how many parameters does it have?
It has a parameter count of 86M. "We trained a large language model with 86M 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.
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.