LightOn Mini

Closed weights LightOn 40B parameters March 2023

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
LightOn
Organisation type
Industry
Country
France
Published
21 March 2023

What it does

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

Domain
Language
Task
Language modeling/generation, Chat
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
40B

"Boasting an impressive 40 billion parameters, Mini is a formidable addition to the growing array of language models available in the market today."

Training data
1,000,000,000,000 tokens

"The amount of data in Mini corpus is 1 trillion tokens. We mainly used data from the public web to pre-train our model, with strong filtering, toxicity reduction, and deduplication to ensure that only high-quality data is retained." assuming 0.75 words per token - 750000000000.0 words

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
2.4 × 10²³ FLOP

6ND aproximation: 6*40B*1T = 2.4e23

How it was established
Operation counting

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
Hosted access (no API)
Training code
Unreleased

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
Record confidence
Confident

Sources

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

Reference
LightOn's Large Language Model of 40 billion parameters: MINI
Last updated
28 November 2025

What the numbers mean

What this model is

LightOn Mini was published by LightOn, in the country recorded as France, during March 2023. It comes out of an organisation categorised as industry.

It works in the domain of Language, and is recorded as performing the task of language modeling/generation, Chat.

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

How it was trained

Training it took a computation budget of roughly 2.4 × 10²³ 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 1,000,000,000,000 tokens of text.

Answers

LightOn Mini — common questions

01

LightOn Mini— how much compute was used to train it?

Training consumed around 2.4 × 10²³ 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.

02

LightOn Mini— 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.

03

LightOn Mini— is it open source?

No. Its weights have not been published, so it exists only as a service controlled by its owner.

04

LightOn Mini— how many parameters does it have?

It has a parameter count of 40B. "Boasting an impressive 40 billion parameters, Mini is a formidable addition to the growing array of language models available in the market today.". 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.

05

LightOn Mini— who created it?

It was published by LightOn, based in France, an organisation categorised as industry.

06

LightOn Mini— when was it released?

It was published in March 2023. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.

07

LightOn Mini— what is it used for?

It works in the domain of Language, and is recorded as handling the task of language modeling/generation, Chat. 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.

Source

Original publication

Record last updated 28 November 2025

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.