RiNALMo

Closed weights University of Zagreb,Genome Institute of Singapore,Bioinformatics Institute 650M parameters February 2024

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 Zagreb,Genome Institute of Singapore,Bioinformatics Institute
Organisation type
Academia,Academia,Government
Country
Croatia, Singapore
Published
29 February 2024
Authors
Rafael Josip Penić, Tin Vlašić, Roland G. Huber, Yue Wan, Mile Šikić

What it does

The problem areas the model was built for. A model can carry several of each.

Domain
Biology
Task
RNA structure prediction, RNA splice-site prediction, Mean ribosome load 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
650M
Training data
tokens

17,000,000 RNA sequences

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 × 10²¹ FLOP

1. Hardware: 7x NVIDIA A100 GPUs (3.12e14 FLOP/s per GPU) 2. Training duration: 2 weeks (directly provided) = 1,209,600 seconds 3. Utilization: 40% (0.4) 4. Calculation: 3.12e14 FLOP/s × 7 GPUs × 1,209,600s × 0.4 = 1.05e21 FLOPs

How it was established
Hardware

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
Chips used
7
Wall-clock time
336 hours (14 days)
Power draw
5.5 kW

How it is classified

Labels the source dataset applies when tracking notable models, and how confident it is in the entry.

Record confidence
Confident
Citations
108

Sources

Where this record came from and when it was last checked.

Reference
RiNALMo: General-Purpose RNA Language Models Can Generalize Well on Structure Prediction Tasks
Last updated
25 May 2026

What the numbers mean

Background

RiNALMo was published by University of Zagreb,Genome Institute of Singapore,Bioinformatics Institute, in Croatia, in February 2024. academia,Academia,Government is the category the publisher falls under.

It works in Biology, and is recorded as doing rNA structure prediction, RNA splice-site prediction, Mean ribosome load prediction.

Its weights were never published, so it can only be reached through its provider. No graphics card changes that.

What went into building it

Producing it required around 1.1 × 10²¹ FLOP of arithmetic, on NVIDIA A100, which is a statement about the training budget rather than about inference.

Answers

RiNALMo — common questions

01

How many parameters does RiNALMo have?

RiNALMo has 650M 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.

02

Who created RiNALMo?

RiNALMo was published by University of Zagreb,Genome Institute of Singapore,Bioinformatics Institute, based in Croatia, categorised as academia,Academia,Government.

03

When was RiNALMo released?

RiNALMo was published in February 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.

04

What is RiNALMo used for?

RiNALMo works in Biology, and is recorded as handling rNA structure prediction, RNA splice-site prediction, Mean ribosome load 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.

05

How much compute was used to train RiNALMo?

Around 1.1 × 10²¹ FLOP, on NVIDIA A100. 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.

06

What GPU do I need to run RiNALMo?

None. RiNALMo 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.

07

Is RiNALMo open source?

The licensing for RiNALMo was never recorded in our source data. We treat unstated licensing as closed, because an unrecorded licence is not one to rely on.

Source

Original publication

Record last updated 25 May 2026

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.