Gemma 4 12B TPS calculator

Open weights Google DeepMind 12B parameters June 2026

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

509 of 818 cards that can run it

Smallest card that fits

Xeon Phi 5110P

8 GB · Q3_K_M · 19.9 tok/s

Fastest card

B200

284 tok/s · 180 GB

Which GPUs can run Gemma 4 12B?

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

509 cards match

Calculating

					Needs	Quantisation	Fit
284 tok/s 170–454 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	13.5 GB	Q8_0	Comfortable
284 tok/s 170–454 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	13.5 GB	Q8_0	Comfortable
226 tok/s 136–362 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	13.5 GB	Q8_0	Comfortable
226 tok/s 136–362 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	13.5 GB	Q8_0	Comfortable
181 tok/s 109–290 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	13.5 GB	Q8_0	Comfortable
173 tok/s 104–277 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	13.5 GB	Q8_0	Comfortable
173 tok/s 104–277 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	13.5 GB	Q8_0	Comfortable
166 tok/s 100–265 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	13.5 GB	Q8_0	Comfortable
147 tok/s 88–236 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	13.5 GB	Q8_0	Comfortable
147 tok/s 88–236 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	13.5 GB	Q8_0	Comfortable
147 tok/s 88–236 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	13.5 GB	Q8_0	Comfortable
143 tok/s 86–228 · low confidence	CMP 170HX 8 GB NVIDIA	8 GB	1,490 GB/s	Sep 2021	6.5 GB	Q3_K_M	Tight
140 tok/s 84–223 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	13.5 GB	Q8_0	Comfortable
128 tok/s 77–204 · low confidence	CMP 170HX 10 GB NVIDIA	10 GB	1,560 GB/s	Sep 2021	7.9 GB	Q4_K_M	Tight
119 tok/s 71–191 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	13.5 GB	Q8_0	Comfortable
119 tok/s 71–191 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	13.5 GB	Q8_0	Comfortable
119 tok/s 71–191 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	13.5 GB	Q8_0	Comfortable
119 tok/s 71–191 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	13.5 GB	Q8_0	Comfortable
119 tok/s 71–191 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	13.5 GB	Q8_0	Comfortable
90.7 tok/s 54–145 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	13.5 GB	Q8_0	Comfortable
90.7 tok/s 54–145 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	13.5 GB	Q8_0	Comfortable
75.6 tok/s 45–121 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	13.5 GB	Q8_0	Comfortable
74.0 tok/s 44–118 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	13.5 GB	Q8_0	Comfortable
73.5 tok/s 44–118 · low confidence	RTX A5000-8Q NVIDIA	8 GB	768 GB/s	Apr 2021	6.5 GB	Q3_K_M	Tight
72.3 tok/s 43–116 · low confidence	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	13.5 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 DeepMind
Organisation type: Industry
Country: United States of America
Published: 3 June 2026

What it does

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

Domain: Language, Multimodal, Audio
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: 12B
Training data: tokens

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 (unrestricted)
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.

Last updated: 10 June 2026

The extremes

The ten fastest GPUs for Gemma 4 12B

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 Gemma 4 12B

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

Hardware requirements in practice

Minimum card

Xeon Phi 5110P

Memory needed

6.5 GB

Fastest

284 tok/s

Gemma 4 12B is small enough at 12B parameters that hardware is rarely the obstacle — 509 of the cards we track can run it, including cards several years old.

The entry point is the Xeon Phi 5110P: 8 GB of memory, Q3_K_M compression, roughly 19.9 tokens per second.

The quickest result comes from a B200 at around 284 tokens per second — its 8,000 GB/s of bandwidth is what buys that.

Where it came from

Gemma 4 12B was published by Google DeepMind, in United States of America, in June 2026. It comes out of industry.

It works in Language, Multimodal, Audio, and is recorded as doing language modeling/generation, Question answering.

Because its weights were released, nothing about running it depends on a provider staying available — it is yours once downloaded. It is published under the google organisation on Hugging Face.

Understanding the speeds

Half the cards that hold it manage more than 20.8 tokens per second, and 455 exceed reading speed outright.

Every weight participates in every token here, so bandwidth is the whole story: the ranking below is effectively a ranking of memory throughput.

Without the attention layout on record, the memory column is an approximation. It is close enough to choose hardware by, and least reliable at long context.

Step by step

How to choose a GPU for Gemma 4 12B

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

Look at what Gemma 4 12B actually needs — around 6.5 GB at Q3_K_M. No amount of processing power compensates for a card that cannot hold it.
02

Decide how long your conversations run

Longer conversations cost memory on top of what the weights need. Move the slider to your real working length before trusting any row for Gemma 4 12B.
03

Choose how far you will compress it

Each card runs the least-compressed copy it can hold — Q3_K_M on the smallest card that fits. Setting a floor drops the cards that only manage Gemma 4 12B by squeezing it further than you would want.
04

Compare tokens per second, not specifications

Ranking by tokens per second for Gemma 4 12B follows memory bandwidth, not core counts, which is why the B200 tops it at 284 tok/s.
05

Check the fit verdict before buying

Tight means Gemma 4 12B loads and works, with no room to raise the context later. Comfortable means you can. The difference matters more than a few tokens per second.
06

Check the card from the other side

Each card page repeats this sweep for every model we hold. It answers what else the hardware is good for, beyond Gemma 4 12B.

Answers

Gemma 4 12B — common questions

How much VRAM does Gemma 4 12B need?

About 6.5 GB at Q3_K_M compression, 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.

Can I run Gemma 4 12B on a 8 GB GPU?

Yes. A CMP 170HX 8 GB with 8 GB runs it at Q3_K_M, using about 6.5 GB and generating roughly 143 tokens per second — a tight fit.

Can I run Gemma 4 12B on a 12 GB GPU?

Yes. A GeForce RTX 3080 Ti with 12 GB runs it at Q6_K, using about 10.7 GB and generating roughly 47.0 tokens per second — a tight fit.

Can I run Gemma 4 12B on a 16 GB GPU?

Yes. A Tesla V100 SXM2 16 GB with 16 GB runs it at Q8_0, using about 13.5 GB and generating roughly 40.1 tokens per second — a tight fit.

Can I run Gemma 4 12B on a 24 GB GPU?

Yes. A GeForce RTX 5090 D V2 with 24 GB runs it at Q8_0, using about 13.5 GB and generating roughly 47.5 tokens per second — a comfortable fit.

Is Gemma 4 12B open source?

Its weights are published, so Gemma 4 12B 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.

How many parameters does Gemma 4 12B have?

Gemma 4 12B has 12B parameters. 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.

Who created Gemma 4 12B?

Gemma 4 12B was published by Google DeepMind, based in United States of America, categorised as industry.

When was Gemma 4 12B released?

Gemma 4 12B was published in June 2026.

What is Gemma 4 12B used for?

Gemma 4 12B works in Language, Multimodal, Audio, and is recorded as handling language modeling/generation, Question answering. These are the areas it was designed around; they describe intent rather than a hard boundary.

Where can I download Gemma 4 12B?

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

Can I run Gemma 4 12B if it does not fit in my GPU?

Only by offloading, which is usually a false economy: the part in system memory drags the whole thing down — the nearest miss we calculate is short by 2.5 GB. Our figures for Gemma 4 12B assume it is fully resident.

Would two GPUs run Gemma 4 12B faster?

A second card roughly doubles the memory available but not the generation rate. With 509 cards already able to run Gemma 4 12B alone, the case for pairing is weak.

Why does the quantisation differ between cards for Gemma 4 12B?

Because capacity varies, so does how hard Gemma 4 12B has to be squeezed — 5 distinct levels appear in the table above. Set a minimum quality to compare at one.

How accurate are these Gemma 4 12B speed estimates?

Every figure is derived from memory bandwidth and model size, not benchmarked. That is why each is published as a range such as 170–454 tok/s on the B200 rather than a single number.

What GPU do I need to run Gemma 4 12B?

The smallest card in our catalogue that holds Gemma 4 12B is the Xeon Phi 5110P, with 8 GB of memory. It runs the model at Q3_K_M using about 6.5 GB, and produces roughly 19.9 tokens per second. 509 cards in total can run it.

How fast is Gemma 4 12B on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 284 tokens per second; the slowest that still runs it manages considerably less. Reading speed is around ten tokens per second, and 455 of the cards that can run Gemma 4 12B clear that.

Source

Original publication

Record last updated 10 June 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