OpenDiLoCo 150M
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
- Prime Intellect
- Organisation type
- Industry
- Country
- United States of America
- Published
- 10 July 2024
- Authors
- Sami Jaghouar, Jack Min Ong, Johannes Hagemann
What it does
The problem areas the model was built for. A model can carry several of each.
- Domain
- Language
- Task
- Language modeling/generation, Question answering
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
- 150M
- Training data
- tokens
"We conduct various experiments using a model with 150 million parameters on a language modeling task using the C4 dataset (Raffel et al., 2019). "
"The hyperparameters are consistent with DiLoCoacross experiments: an inner learning rate of 4e−4, 1,000 warm-up steps, 0.1 weight decay, a batch size of 512, a sequence length of 1,024, a learning rate for the Nesterov outer optimizer of 0.7, and Nesterov momentum of 0.9. Similarly, we run the experiments for a total of 88,000 steps." 512*1024*88000=46137344000
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
- 4.2 × 10¹⁹ FLOP
- How it was established
- Operation counting
6ND = 6 FLOP / token / parameter * 150*10^6 parameters * 46137344000 tokens [see dataset size notes] = 4.152361e+19 FLOP
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
https://github.com/PrimeIntellect-ai/OpenDiLoCo Apache 2.0
How it is classified
Labels the source dataset applies when tracking notable models, and how confident it is in the entry.
- Record confidence
- Confident
Sources
Where this record came from and when it was last checked.
- Reference
- OpenDiLoCo: An Open-Source Framework for Globally Distributed Low-Communication Training
- Last updated
- 28 November 2025
What the numbers mean
Background
OpenDiLoCo 150M was published by Prime Intellect, in the country recorded as United States of America, during July 2024. The category the publisher falls under is industry.
It works in the domain of Language, and is recorded as performing the task of language modeling/generation, Question answering.
Because the weights are not available, none of the hardware figures elsewhere on this site apply to it.
How it was trained
Training it took a computation budget of roughly 4.2 × 10¹⁹ FLOP. That figure measures what producing the model cost, and has no bearing on how fast it answers.
Answers
OpenDiLoCo 150M — common questions
OpenDiLoCo 150M— 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.
OpenDiLoCo 150M— is it open source?
No. Its weights have not been published, so it exists only as a service controlled by its owner.
OpenDiLoCo 150M— how many parameters does it have?
It has a parameter count of 150M. "We conduct various experiments using a model with 150 million parameters on a language modeling task using the C4 dataset (Raffel et al., 2019). ". 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.
OpenDiLoCo 150M— who created it?
It was published by Prime Intellect, based in United States of America, an organisation categorised as industry.
OpenDiLoCo 150M— when was it released?
It was published in July 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.
OpenDiLoCo 150M— what is it used for?
It works in the domain of Language, and is recorded as handling the task of language modeling/generation, Question answering. 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.
OpenDiLoCo 150M— how much compute was used to train it?
Training consumed around 4.2 × 10¹⁹ 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.
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.