PaliGemma 2 3B Mix 448 TPS calculator

Open weights Google 2.9B parameters February 2025

Each card below is assessed against this model at the context length and minimum quality you choose. Speed is an estimate for a single request, calculated from the card's memory bandwidth and the size of the model once compressed.

Calculated for this model

818 cards that can run it

818 cards we hold specifications for

Smallest card that fits

Tesla C1080

4 GB · Q6_K · 18.3 tok/s

Fastest card

B200

1,160 tok/s · 180 GB

Which GPUs can run PaliGemma 2 3B Mix 448?

Set the inputs, read the answer

Context length tokens

A longer conversation needs more memory, which can push this model off smaller cards.

Minimum quality

Hides cards that would only fit the model by compressing it below this point.

Search cards

Card memory

Also show cards that cannot run it

818 cards match

Calculating

					Needs	Quantisation	Fit
1,160 tok/s 696–1,857 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	3.8 GB	Q8_0	Comfortable
1,160 tok/s 696–1,857 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	3.8 GB	Q8_0	Comfortable
927 tok/s 556–1,483 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	3.8 GB	Q8_0	Comfortable
927 tok/s 556–1,483 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	3.8 GB	Q8_0	Comfortable
741 tok/s 445–1,186 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	3.8 GB	Q8_0	Comfortable
709 tok/s 426–1,135 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	3.8 GB	Q8_0	Comfortable
709 tok/s 426–1,135 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	3.8 GB	Q8_0	Comfortable
679 tok/s 407–1,086 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	3.8 GB	Q8_0	Comfortable
602 tok/s 361–964 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	3.8 GB	Q8_0	Comfortable
602 tok/s 361–964 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	3.8 GB	Q8_0	Comfortable
602 tok/s 361–964 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	3.8 GB	Q8_0	Comfortable
571 tok/s 343–914 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	3.8 GB	Q8_0	Comfortable
487 tok/s 292–780 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	3.8 GB	Q8_0	Comfortable
487 tok/s 292–780 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	3.8 GB	Q8_0	Comfortable
487 tok/s 292–780 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	3.8 GB	Q8_0	Comfortable
487 tok/s 292–780 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	3.8 GB	Q8_0	Comfortable
487 tok/s 292–780 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	3.8 GB	Q8_0	Comfortable
371 tok/s 223–594 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	3.8 GB	Q8_0	Comfortable
371 tok/s 223–594 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	3.8 GB	Q8_0	Comfortable
309 tok/s 186–495 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	3.8 GB	Q8_0	Comfortable
303 tok/s 182–484 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	3.8 GB	Q8_0	Comfortable
296 tok/s 178–473 · low confidence	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	3.8 GB	Q8_0	Comfortable
296 tok/s 178–473 · low confidence	A100X NVIDIA	80 GB	2,040 GB/s	Jun 2021	3.8 GB	Q8_0	Comfortable
296 tok/s 178–473 · low confidence	A800 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Aug 2022	3.8 GB	Q8_0	Comfortable
296 tok/s 178–473 · low confidence	H100 CNX NVIDIA	80 GB	2,040 GB/s	Mar 2023	3.8 GB	Q8_0	Comfortable

Speeds are estimates for a single request — one conversation at a time — calculated from memory bandwidth, model size and quantisation. Real throughput varies with the inference runtime and its version. Figures published by hardware vendors measure many simultaneous requests and are much higher.

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
Organisation type: Industry
Country: United States of America
Published: 19 February 2025
Authors: Omar Sanseviero, Andreas Steiner

What it does

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

Domain: Vision, Multimodal, Language
Task: Image captioning, Visual question answering, Object detection, Object recognition, Image segmentation, Character recognition (OCR)
Base model: PaliGemma 2 3B Mix 224

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: 2.9B
Training data: 51,200,000,000 tokens

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 v5e
Chips used: 256
Power draw: 113.1 kW

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: Open — downloadable
Model access: Open weights (non-commercial)
Training code: Unreleased
Hugging Face: google

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: Introducing PaliGemma 2 mix: A vision-language model for multiple tasks
Last updated: 28 November 2025

The extremes

The ten fastest GPUs that run PaliGemma 2 3B Mix 448

Ranked by estimated tokens per second, newest card first where speeds tie. Because generation is bound by memory bandwidth, this ordering follows bandwidth rather than any gaming benchmark.

The smallest GPUs that still run PaliGemma 2 3B Mix 448

The cheapest route in, by memory capacity. A tight fit runs the model but leaves nothing spare for a longer conversation.

What the numbers mean

What you need to run it

Minimum card

Tesla C1080

Memory needed

3.1 GB

Fastest

1,160 tok/s

PaliGemma 2 3B Mix 448 reaches a parameter count of 2.9B. That is small enough that hardware is rarely the obstacle, including on cards several years old. The number of cards we track that can run it: 818.

The least hardware that works is Tesla C1080, with a memory capacity of 4 GB, running it at a compression of Q6_K and producing around 18.3 tokens per second.

At the other end sits B200, generating roughly 1,160 tokens per second on the strength of a memory bandwidth of 8,000 GB/s.

Where it came from

PaliGemma 2 3B Mix 448 was published by Google, in the country recorded as United States of America, during February 2025. The publishing organisation is categorised as industry.

It works in the domain of Vision, Multimodal, Language, and is recorded as performing the task of image captioning, Visual question answering, Object detection, Object recognition, Image segmentation, Character recognition (OCR).

It builds on PaliGemma 2 3B Mix 224. That is why it shares the base model's general shape and size.

Published weights mean the model runs on your machine rather than someone else's, which is what makes the hardware question below answerable at all. On Hugging Face it is published under the organisation google.

Understanding the speeds

Half the cards that hold it manage more than 36.8 tokens per second. Exceeding reading speed outright: 783 of them.

It is a dense model, so every parameter is read for every token produced. That makes speed track memory bandwidth almost exactly — a card with twice the bandwidth generates roughly twice as fast.

Memory here is estimated from size rather than computed from the architecture, which is not recorded for this model — the numbers are indicative rather than exact.

What went into building it

It was trained on a corpus of about 51,200,000,000 tokens of text.

Step by step

How to choose a GPU for PaliGemma 2 3B Mix 448

The table above has already assessed every card we hold specifications for against this model. Getting to your answer takes six steps.

01

Start from the memory column

Start from what it actually needs, which is the requirement of PaliGemma 2 3B Mix 448, needing around 3.1 GB at a compression of Q6_K. That figure, not the headline performance of a card, is what decides whether it runs.
02

Decide how long your conversations run

Set the context to what you will actually use. The cache grows with the conversation, and it is the usual reason a card that seemed fine stops fitting PaliGemma 2 3B Mix 448.
03

Set a quality floor

Each card runs the least-compressed copy it can hold, reaching a compression of Q6_K on the smallest card that fits. Setting a minimum quality drops the cards that only manage it by squeezing further than you would want, and holds the comparison at one level.
04

Sort by speed

The speed ordering is effectively an ordering by memory bandwidth, for PaliGemma 2 3B Mix 448. It will not match a gaming ordering, because generation is bound by memory bandwidth. The card topping the list is B200, at 1,160 tok/s.
05

Check the fit verdict before buying

The fit column separates cards that just manage it from those with room to spare, in the case of PaliGemma 2 3B Mix 448. Comfortable means you can grow the context later. That difference matters more than a few tokens per second, so buy for comfortable if you expect to.
06

See what else that card runs

Following a card through to its own page shows every other model it can hold, which is the question that follows once you have settled on PaliGemma 2 3B Mix 448.

Answers

PaliGemma 2 3B Mix 448 — common questions

PaliGemma 2 3B Mix 448— can I run it on a GPU holding 24 GB?

Yes. The card GeForce RTX 5090 D V2, holding 24 GB, runs it at a compression of Q8_0, using about 3.8 GB and generating roughly 194 tokens per second. The fit is comfortable.

PaliGemma 2 3B Mix 448— is it open source?

Its weights are published, so it can be downloaded and run on your own hardware. Note that open weights is not the same as open source in the full sense — it says nothing about the training data, the training code, or the commercial terms attached.

PaliGemma 2 3B Mix 448— how many parameters does it have?

It has a parameter count of 2.9B. https://huggingface.co/google/paligemma-3b-mix-448. 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.

PaliGemma 2 3B Mix 448— who created it?

It was published by Google, based in United States of America, an organisation categorised as industry.

PaliGemma 2 3B Mix 448— when was it released?

It was published in February 2025.

PaliGemma 2 3B Mix 448— what is it used for?

It works in the domain of Vision, Multimodal, Language, and is recorded as handling the task of image captioning, Visual question answering, Object detection, Object recognition, Image segmentation, Character recognition (OCR). Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.

PaliGemma 2 3B Mix 448— where can I download it?

Its weights are published on Hugging Face, under the organisation google. We do not host model files — this site calculates what hardware is needed to run them.

PaliGemma 2 3B Mix 448— can I run it if it does not fit in my GPU?

Only by offloading, which is usually a false economy: the part held in system memory drags the whole thing down. Every figure here assumes the whole model is resident on the card.

PaliGemma 2 3B Mix 448— would two GPUs run it faster?

Capacity adds across cards; throughput does not. The number of cards already holding it on their own: 818. So a second card is rarely the answer here.

PaliGemma 2 3B Mix 448— why does the quantisation differ between cards?

A larger card holds a more accurate copy. The number of compression levels used across the cards that run it: 2. Bigger cards get the more accurate version, and the quality floor above pins the comparison to one level.

PaliGemma 2 3B Mix 448— how accurate are these speed estimates?

They are calculated from specifications rather than measured, and each carries a range. One example: 696–1,857 tok/s on B200. The same model and card vary by thirty to fifty per cent depending on the inference software and its version.

PaliGemma 2 3B Mix 448— what GPU do I need to run it?

The smallest card in our catalogue that holds it is Tesla C1080, with a memory capacity of 4 GB. It runs the model at a compression of Q6_K using about 3.1 GB, and produces roughly 18.3 tokens per second. The number of cards able to run it in total: 818.

PaliGemma 2 3B Mix 448— how fast is it on a GPU?

It depends on the card. The quickest we calculate is B200, at about 1,160 tokens per second; the slowest that still runs it manages considerably less. Reading speed is around ten tokens per second, and the number of cards clearing that: 783.

PaliGemma 2 3B Mix 448— how much VRAM does it need?

It needs about 3.1 GB at a compression of Q6_K, which is what the smallest card that runs it uses. Less compression needs more: the figures in the memory column above are recalculated for each card, because each one holds the least-compressed version it can.

PaliGemma 2 3B Mix 448— can I run it on a GPU holding 8 GB?

Yes. The card CMP 170HX 8 GB, holding 8 GB, runs it at a compression of Q8_0, using about 3.8 GB and generating roughly 216 tokens per second. The fit is comfortable.

PaliGemma 2 3B Mix 448— can I run it on a GPU holding 12 GB?

Yes. The card GeForce RTX 3080 Ti, holding 12 GB, runs it at a compression of Q8_0, using about 3.8 GB and generating roughly 132 tokens per second. The fit is comfortable.

PaliGemma 2 3B Mix 448— can I run it on a GPU holding 16 GB?

Yes. The card Tesla V100 SXM2 16 GB, holding 16 GB, runs it at a compression of Q8_0, using about 3.8 GB and generating roughly 164 tokens per second. The fit is comfortable.

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.

All AI models