T5Gemma (Gemma 9B-9B) TPS calculator

Open weights Google DeepMind 16.7B parameters April 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

306 cards that can run it

818 cards we hold specifications for

Smallest card that fits

P102-101

10 GB · Q3_K_M · 18.6 tok/s

Fastest card

B200

203 tok/s · 180 GB

Which GPUs can run T5Gemma (Gemma 9B-9B)?

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

306 cards match

Calculating

					Needs	Quantisation	Fit
203 tok/s 122–325 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	18.6 GB	Q8_0	Comfortable
203 tok/s 122–325 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	18.6 GB	Q8_0	Comfortable
162 tok/s 97–259 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	18.6 GB	Q8_0	Comfortable
162 tok/s 97–259 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	18.6 GB	Q8_0	Comfortable
130 tok/s 78–207 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	18.6 GB	Q8_0	Comfortable
124 tok/s 74–198 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	18.6 GB	Q8_0	Comfortable
124 tok/s 74–198 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	18.6 GB	Q8_0	Comfortable
119 tok/s 71–190 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	18.6 GB	Q8_0	Comfortable
107 tok/s 64–171 · low confidence	CMP 170HX 10 GB NVIDIA	10 GB	1,560 GB/s	Sep 2021	8.9 GB	Q3_K_M	Tight
105 tok/s 63–169 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	18.6 GB	Q8_0	Comfortable
105 tok/s 63–169 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	18.6 GB	Q8_0	Comfortable
105 tok/s 63–169 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	18.6 GB	Q8_0	Comfortable
99.9 tok/s 60–160 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	18.6 GB	Q8_0	Comfortable
85.2 tok/s 51–136 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	18.6 GB	Q8_0	Comfortable
85.2 tok/s 51–136 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	18.6 GB	Q8_0	Comfortable
85.2 tok/s 51–136 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	18.6 GB	Q8_0	Comfortable
85.2 tok/s 51–136 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	18.6 GB	Q8_0	Comfortable
85.2 tok/s 51–136 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	18.6 GB	Q8_0	Comfortable
64.9 tok/s 39–104 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	18.6 GB	Q8_0	Comfortable
64.9 tok/s 39–104 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	18.6 GB	Q8_0	Comfortable
56.8 tok/s 34–91 · low confidence	GeForce RTX 3080 12 GB NVIDIA	12 GB	912 GB/s	Jan 2022	9.8 GB	IQ4_XS	Tight
56.8 tok/s 34–91 · low confidence	GeForce RTX 3080 Ti NVIDIA	12 GB	912 GB/s	May 2021	9.8 GB	IQ4_XS	Tight
54.1 tok/s 32–87 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	18.6 GB	Q8_0	Comfortable
52.9 tok/s 32–85 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	18.6 GB	Q8_0	Comfortable
52.0 tok/s 31–83 · low confidence	CMP 90HX NVIDIA	10 GB	760 GB/s	Jul 2021	8.9 GB	Q3_K_M	Tight

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 DeepMind
Organisation type: Industry
Country: United States of America
Published: 8 April 2025
Authors: Biao Zhang, Fedor Moiseev, Joshua Ainslie, Paul Suganthan, Min Ma, Surya Bhupatiraju, Fede Lebron, Orhan Firat, Armand Joulin, Zhe Dong

What it does

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

Domain: Language
Task: Semantic embedding, Language modeling/generation, Question answering
Base model: Gemma 2 9B

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: 16.7B
Training data: tokens

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.

How it was established: Operation counting
Fine-tuning compute: 2 × 10²³ 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: Open — downloadable
Model access: Open weights (restricted use)
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: Likely

Sources

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

Reference: Encoder-Decoder Gemma: Improving the Quality-Efficiency Trade-Off via Adaptation
Last updated: 11 February 2026

The extremes

The ten fastest GPUs that run T5Gemma (Gemma 9B-9B)

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 T5Gemma (Gemma 9B-9B)

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

The hardware side

Minimum card

P102-101

Memory needed

8.9 GB

Fastest

203 tok/s

T5Gemma (Gemma 9B-9B) reaches a parameter count of 16.7B. 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: 306.

At the low end it is handled by P102-101, with a memory capacity of 10 GB, running it at a compression of Q3_K_M and producing around 18.6 tokens per second.

The fastest we calculate for it is B200, generating roughly 203 tokens per second on the strength of a memory bandwidth of 8,000 GB/s.

Where it came from

T5Gemma (Gemma 9B-9B) was published by Google DeepMind, in the country recorded as United States of America, during April 2025. The category the publisher falls under is industry.

It works in the domain of Language, and is recorded as performing the task of semantic embedding, Language modeling/generation, Question answering.

Rather than being trained from scratch, it is derived from Gemma 2 9B. 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

Across every card that can run it, the middle of the range sits at 20.9 tokens per second. Exceeding reading speed outright: 268 of them.

Being dense, it reads all of itself per token, which is why the ordering by speed below follows the ordering by memory bandwidth so closely.

Its internal architecture is not on file, so memory is approximated from the parameter count and marked accordingly. Expect the real figure to differ, more so at long context.

Step by step

How to choose a GPU for T5Gemma (Gemma 9B-9B)

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

01

Read the memory figure first

The table lists every card able to hold T5Gemma (Gemma 9B-9B), needing around 8.9 GB at a compression of Q3_K_M. Capacity is the gate — a card either holds it or it does not.
02

Set the context length you will work at

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 T5Gemma (Gemma 9B-9B).
03

Choose how far you will compress it

Each card runs the least-compressed copy it can hold, reaching a compression of Q3_K_M 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

Compare tokens per second, not specifications

Sort by speed to see how cards rank for T5Gemma (Gemma 9B-9B). It will not match a gaming ordering, because generation is bound by memory bandwidth. The card topping the list is B200, at 203 tok/s.
05

Check the fit verdict before buying

Tight means it loads and works with no room to raise the context later, in the case of T5Gemma (Gemma 9B-9B). 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

Open the card you have settled on

Every card name links to its own page, which runs the same calculation across the whole model catalogue. A card is usually bought for more than one model, so it is worth a look before buying for T5Gemma (Gemma 9B-9B).

Answers

T5Gemma (Gemma 9B-9B) — common questions

T5Gemma (Gemma 9B-9B)— would two GPUs run it faster?

Two cards buy memory rather than speed, which matters only if one card cannot hold it. The number that can: 306. So a second card is rarely the answer here.

T5Gemma (Gemma 9B-9B)— why does the quantisation differ between cards?

Each card is shown running the least-compressed copy it can hold. The number of distinct compression levels across the cards that fit it: 5. Bigger cards get the more accurate version, and the quality floor above pins the comparison to one level.

T5Gemma (Gemma 9B-9B)— how accurate are these speed estimates?

They are calculated from specifications rather than measured, and each carries a range. One example: 122–325 tok/s on B200. The same model and card vary by thirty to fifty per cent depending on the inference software and its version.

T5Gemma (Gemma 9B-9B)— what GPU do I need to run it?

The smallest card in our catalogue that holds it is P102-101, with a memory capacity of 10 GB. It runs the model at a compression of Q3_K_M using about 8.9 GB, and produces roughly 18.6 tokens per second. The number of cards able to run it in total: 306.

T5Gemma (Gemma 9B-9B)— how fast is it on a GPU?

It depends on the card. The quickest we calculate is B200, at about 203 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: 268.

T5Gemma (Gemma 9B-9B)— how much VRAM does it need?

It needs about 8.9 GB at a compression of Q3_K_M, 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.

T5Gemma (Gemma 9B-9B)— 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 IQ4_XS, using about 9.8 GB and generating roughly 56.8 tokens per second. The fit is tight.

T5Gemma (Gemma 9B-9B)— 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 Q5_K_M, using about 12.7 GB and generating roughly 51.2 tokens per second. The fit is tight.

T5Gemma (Gemma 9B-9B)— 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 18.6 GB and generating roughly 34.0 tokens per second. The fit is tight.

T5Gemma (Gemma 9B-9B)— 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.

T5Gemma (Gemma 9B-9B)— how many parameters does it have?

It has a parameter count of 16.7B. 16.7B. 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.

T5Gemma (Gemma 9B-9B)— who created it?

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

T5Gemma (Gemma 9B-9B)— when was it released?

It was published in April 2025.

T5Gemma (Gemma 9B-9B)— what is it used for?

It works in the domain of Language, and is recorded as handling the task of semantic embedding, Language modeling/generation, Question answering. These are the areas it was designed around; they describe intent rather than a hard boundary.

T5Gemma (Gemma 9B-9B)— 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.

T5Gemma (Gemma 9B-9B)— can I run it if it does not fit in my GPU?

Partly. Layers that do not fit sit in system memory and run at a fraction of the speed, so a mostly-offloaded model is rarely worth using. The nearest miss we calculate falls short by 3.6 GB. Every figure here assumes the whole model is resident on the card.

Source

Original publication

Record last updated 11 February 2026

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