LIMoE-H/14
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
- Organisation type
- Industry
- Country
- United States of America
- Published
- 6 June 2022
- Authors
- Basil Mustafa, Carlos Riquelme, Joan Puigcerver, Rodolphe Jenatton, Neil Houlsby
What it does
The problem areas the model was built for. A model can carry several of each.
- Domain
- Multimodal, Vision, Language
- Task
- Image classification
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
- 5.6B
- Training data
- 4,575,625,612,000 tokens
Section 1: "We scale this up to a large 5.6B parameter LIMoE-H/14"
"By default, all models are trained on paired image-text data used in [16], consisting of 3.6B images and alt-texts scraped from the web. For large LIMoE-H/14 experiment, we also co-train with JFT-4B [17]." "The largest scale model is trained at batch size 21502, with resolution 288 and text sequence length 16." "The model was trained for 700k steps pre-cooldown. There was one cooldown of length 125k steps from the final step, and 3 of length 40k steps starting from step 650k" 21502*288*70000…
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
- 1.8 × 10²² FLOP
- How it was established
- Operation counting
Section 3.2: "The model contains 5.6B parameters in total, but only applies 675M parameters per token" From Section A.3, "batch size 21502 with resolution 288 and text sequence length16". "The model was trained for 700k steps pre-cooldown. There was one cooldown of length 125k steps from the final step, and 3 of length 40k steps starting from step 650k". Patch size 14 for images. Assume C = 6*N*D. C = 6*675e6*21.5e3*1e6*(16+(288/14)**2)/2 = 1.8e22 This is broadly consistent with ViT-H/14's c…
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
- Google TPU v2,Google TPU v3,Google TPU v4
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
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
- 317
Sources
Where this record came from and when it was last checked.
- Reference
- Multimodal Contrastive Learning with LIMoE: the Language-Image Mixture of Experts
- Last updated
- 25 May 2026
What the numbers mean
About this model
LIMoE-H/14 was published by Google, in the country recorded as United States of America, during June 2022. The publishing organisation is categorised as industry.
It works in the domain of Multimodal, Vision, Language, and is recorded as performing the task of image classification.
Because the weights are not available, none of the hardware figures elsewhere on this site apply to it.
Training and provenance
Training it took a computation budget of roughly 1.8 × 10²² FLOP, on hardware recorded as Google TPU v2,Google TPU v3,Google TPU v4. That figure measures what producing the model cost, and has no bearing on how fast it answers.
The training set ran to roughly 4,575,625,612,000 tokens of text.
Answers
LIMoE-H/14 — common questions
LIMoE-H/14— when was it released?
It was published in June 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.
LIMoE-H/14— what is it used for?
It works in the domain of Multimodal, Vision, Language, and is recorded as handling the task of image classification. Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.
LIMoE-H/14— how much compute was used to train it?
Training consumed around 1.8 × 10²² FLOP, on hardware recorded as Google TPU v2,Google TPU v3,Google TPU v4. 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.
LIMoE-H/14— 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.
LIMoE-H/14— is it open source?
No. Its weights have not been published, so it exists only as a service controlled by its owner.
LIMoE-H/14— how many parameters does it have?
It has a parameter count of 5.6B. Section 1: "We scale this up to a large 5.6B parameter LIMoE-H/14". 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.
LIMoE-H/14— who created it?
It was published by Google, based in United States of America, an organisation categorised as industry.
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.