PaLI-3

Closed weights Google DeepMind,Google Research,Google Cloud 5B parameters October 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
Google DeepMind,Google Research,Google Cloud
Organisation type
Industry,Industry,Industry
Country
United States of America
Published
17 October 2023
Authors
Xi Chen, Xiao Wang, Lucas Beyer, Alexander Kolesnikov, Jialin Wu, Paul Voigtlaender, Basil Mustafa, Sebastian Goodman, Ibrahim Alabdulmohsin, Piotr Padlewski, Daniel Salz, Xi Xiong, Daniel Vlasic, Filip Pavetic, Keran Rong, Tianli Yu, Daniel Keysers, Xiaohua Zhai, Radu Soricut

What it does

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

Domain
Multimodal, Language, Vision
Task
Visual question answering, Character recognition (OCR), Image captioning
Base model
UL2

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
5B

5B

Training data
tokens

contrastive image pretraining at 224² then high‑res 812–1064²

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.

Why it is tracked
SOTA improvement

Table 2, Table 5

Record confidence
Confident

Sources

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

Reference
PaLI-3 Vision Language Models: Smaller, Faster, Stronger
Last updated
28 November 2025

What the numbers mean

Where it came from

PaLI-3 was published by Google DeepMind,Google Research,Google Cloud, in United States of America, in October 2023. The organisation is categorised as industry,Industry,Industry.

It works in Multimodal, Language, Vision, and is recorded as doing visual question answering, Character recognition (OCR), Image captioning.

It is derived from UL2 rather than trained from scratch, which is the usual way a specialised model is produced.

This is a closed model: the trained values stayed with whoever produced them, and there is no local version to run.

Training and provenance

It is tracked in the underlying dataset for one reason in particular: sOTA improvement.

Answers

PaLI-3 — common questions

01

When was PaLI-3 released?

PaLI-3 was published in October 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.

02

What is PaLI-3 used for?

PaLI-3 works in Multimodal, Language, Vision, and is recorded as handling visual question answering, Character recognition (OCR), Image captioning. 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.

03

What GPU do I need to run PaLI-3?

None. PaLI-3 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.

04

Is PaLI-3 open source?

No. PaLI-3 has not had its weights published, so it exists only as a service controlled by its owner.

05

How many parameters does PaLI-3 have?

PaLI-3 has 5B parameters. 5B. 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.

06

Who created PaLI-3?

PaLI-3 was published by Google DeepMind,Google Research,Google Cloud, based in United States of America, categorised as industry,Industry,Industry.

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.