SEA-LION V3 Gemma2 9B TPS calculator

Open weights AI Singapore 9B parameters December 2024

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 cards that can run it

818 cards we hold specifications for

Smallest card that fits

Xeon Phi 5110P

8 GB · IQ4_XS · 24.0 tok/s

Fastest card

B200

376 tok/s · 180 GB

Which GPUs can run SEA-LION V3 Gemma2 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

509 cards match

Calculating

					Needs	Quantisation	Fit
376 tok/s 320–452	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	11.7 GB	Q8_0	Comfortable
376 tok/s 320–452	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	11.7 GB	Q8_0	Comfortable
301 tok/s 180–481 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	11.7 GB	Q8_0	Comfortable
301 tok/s 180–481 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	11.7 GB	Q8_0	Comfortable
240 tok/s 144–385 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	11.7 GB	Q8_0	Comfortable
230 tok/s 196–276	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	11.7 GB	Q8_0	Comfortable
230 tok/s 196–276	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	11.7 GB	Q8_0	Comfortable
220 tok/s 132–352 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	11.7 GB	Q8_0	Comfortable
195 tok/s 117–313 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	11.7 GB	Q8_0	Comfortable
195 tok/s 117–313 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	11.7 GB	Q8_0	Comfortable
195 tok/s 117–313 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	11.7 GB	Q8_0	Comfortable
185 tok/s 158–222	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	11.7 GB	Q8_0	Comfortable
172 tok/s 146–207	CMP 170HX 8 GB NVIDIA	8 GB	1,490 GB/s	Sep 2021	7.0 GB	IQ4_XS	Tight
158 tok/s 134–190	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	11.7 GB	Q8_0	Comfortable
158 tok/s 134–190	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	11.7 GB	Q8_0	Comfortable
158 tok/s 134–190	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	11.7 GB	Q8_0	Comfortable
158 tok/s 134–190	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	11.7 GB	Q8_0	Comfortable
158 tok/s 134–190	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	11.7 GB	Q8_0	Comfortable
131 tok/s 111–157	CMP 170HX 10 GB NVIDIA	10 GB	1,560 GB/s	Sep 2021	8.6 GB	Q5_K_M	Tight
120 tok/s 72–193 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	11.7 GB	Q8_0	Comfortable
120 tok/s 72–193 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	11.7 GB	Q8_0	Comfortable
100 tok/s 60–161 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	11.7 GB	Q8_0	Comfortable
98.2 tok/s 59–157 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	11.7 GB	Q8_0	Comfortable
96.0 tok/s 82–115	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	11.7 GB	Q8_0	Comfortable
96.0 tok/s 82–115	A100X NVIDIA	80 GB	2,040 GB/s	Jun 2021	11.7 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: AI Singapore
Organisation type: Government
Country: Singapore
Published: 19 December 2024
Authors: AI Singapore

What it does

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

Domain: Language
Task: Question answering, Chat, Language modeling/generation
Approach: Self-supervised learning
Base model: Gemma 2 9B
Numerical format: BF16

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: 9B
Training data: 200,000,000,000 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.

Training compute: 4.5 × 10²³ FLOP
How it was established: Hardware,Operation counting
Fine-tuning compute: 1.3 × 10²² FLOP

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: NVIDIA H100 SXM5 80GB
Chips used: 64
Wall-clock time: 240 hours (10 days)
Power draw: 88.1 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 (restricted use)
Training code: Open source
Hugging Face: aisingapore

How it is classified

Labels the source dataset applies when tracking notable models, and how confident it is in the entry.

Likely above 10²³ FLOP: Yes
Record confidence: Confident

Sources

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

Reference: SEA-LION V3
Last updated: 28 November 2025

The extremes

The ten fastest GPUs that run SEA-LION V3 Gemma2 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 SEA-LION V3 Gemma2 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

Hardware requirements in practice

Minimum card

Xeon Phi 5110P

Memory needed

7.0 GB

Fastest

376 tok/s

SEA-LION V3 Gemma2 9B is small enough at 9B 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, IQ4_XS compression, roughly 24.0 tokens per second.

At the other end, a B200 generates roughly 376 tokens per second on it, on the strength of 8,000 GB/s of memory bandwidth.

About this model

SEA-LION V3 Gemma2 9B was published by AI Singapore, in Singapore, in December 2024. government is the category the publisher falls under.

It works in Language, and is recorded as doing question answering, Chat, Language modeling/generation.

Its starting point was Gemma 2 9B — most models at this scale are adapted from an existing base rather than built from nothing.

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. It is published under the aisingapore organisation on Hugging Face.

How fast it runs, and why

Half the cards that hold it manage more than 26.0 tokens per second, and 473 exceed reading speed outright.

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 attention layout is on file, so the memory figures are computed exactly rather than approximated.

How it was trained

Producing it required around 4.5 × 10²³ FLOP of arithmetic, on NVIDIA H100 SXM5 80GB, which is a statement about the training budget rather than about inference.

The training set ran to roughly 200,000,000,000 tokens.

Step by step

How to choose a GPU for SEA-LION V3 Gemma2 9B

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

The table lists every card that can hold SEA-LION V3 Gemma2 9B — around 7.0 GB at IQ4_XS. That figure, not the card's headline performance, is what decides whether it runs.
02

Set the context length you will work at

The conversation occupies memory too, and grows as it goes. Set the slider to the length you expect: at long context SEA-LION V3 Gemma2 9B can slip off a card that handles short questions easily.
03

Set a quality floor

Compression is what makes SEA-LION V3 Gemma2 9B fit smaller cards, at some cost in accuracy — IQ4_XS on the smallest card that fits. A minimum quality removes the ones that go too far.
04

Rank by throughput rather than spec sheet

Sort by speed to see how cards rank for SEA-LION V3 Gemma2 9B. It will not match a gaming ordering — generation is bound by memory bandwidth, which is why the B200 tops it at 376 tok/s.
05

Check the fit verdict before buying

A tight fit runs SEA-LION V3 Gemma2 9B but leaves nothing spare for a longer conversation; comfortable has headroom. If you expect to grow the context, buy for comfortable.
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 SEA-LION V3 Gemma2 9B is settled.

Answers

SEA-LION V3 Gemma2 9B — common questions

Can I run SEA-LION V3 Gemma2 9B on a 24 GB GPU?

Yes. A GeForce RTX 5090 D V2 with 24 GB runs it at Q8_0, using about 11.7 GB and generating roughly 63.1 tokens per second — a comfortable fit.

Is SEA-LION V3 Gemma2 9B open source?

Its weights are published, so SEA-LION V3 Gemma2 9B 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 SEA-LION V3 Gemma2 9B have?

SEA-LION V3 Gemma2 9B has 9B 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 SEA-LION V3 Gemma2 9B?

SEA-LION V3 Gemma2 9B was published by AI Singapore, based in Singapore, categorised as government.

When was SEA-LION V3 Gemma2 9B released?

SEA-LION V3 Gemma2 9B was published in December 2024. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.

What is SEA-LION V3 Gemma2 9B used for?

SEA-LION V3 Gemma2 9B works in Language, and is recorded as handling question answering, Chat, Language modeling/generation. 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.

Where can I download SEA-LION V3 Gemma2 9B?

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

How much compute was used to train SEA-LION V3 Gemma2 9B?

Around 4.5 × 10²³ FLOP, on NVIDIA H100 SXM5 80GB. 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.

Can I run SEA-LION V3 Gemma2 9B if it does not fit in my GPU?

It can be split between the card and system memory, but SEA-LION V3 Gemma2 9B generates painfully slowly that way — the nearest miss we calculate is short by 2.2 GB. Nothing on this page assumes offloading.

Would two GPUs run SEA-LION V3 Gemma2 9B faster?

Capacity adds across cards; throughput does not. Since 509 of the cards we track already hold SEA-LION V3 Gemma2 9B on their own, a second card is rarely the answer here.

Why does the quantisation differ between cards for SEA-LION V3 Gemma2 9B?

Each card is shown running the least-compressed copy it can hold, and SEA-LION V3 Gemma2 9B appears at 4 different compression levels across the cards that fit it. Bigger cards get the more accurate version.

How accurate are these SEA-LION V3 Gemma2 9B speed estimates?

These are estimates with real error bars. The fastest result here, 320–452 tok/s on the B200, could reasonably land anywhere in its published range depending on which runtime you use.

What GPU do I need to run SEA-LION V3 Gemma2 9B?

The smallest card in our catalogue that holds SEA-LION V3 Gemma2 9B is the Xeon Phi 5110P, with 8 GB of memory. It runs the model at IQ4_XS using about 7.0 GB, and produces roughly 24.0 tokens per second. 509 cards in total can run it.

How fast is SEA-LION V3 Gemma2 9B on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 376 tokens per second; the slowest that still runs it manages considerably less. Reading speed is around ten tokens per second, and 473 of the cards that can run SEA-LION V3 Gemma2 9B clear that.

How much VRAM does SEA-LION V3 Gemma2 9B need?

About 7.0 GB at IQ4_XS 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 SEA-LION V3 Gemma2 9B on a 8 GB GPU?

Yes. A CMP 170HX 8 GB with 8 GB runs it at IQ4_XS, using about 7.0 GB and generating roughly 172 tokens per second — a tight fit.

Can I run SEA-LION V3 Gemma2 9B on a 12 GB GPU?

Yes. A GeForce RTX 3080 Ti with 12 GB runs it at Q6_K, using about 9.6 GB and generating roughly 62.4 tokens per second — a tight fit.

Can I run SEA-LION V3 Gemma2 9B on a 16 GB GPU?

Yes. A Tesla V100 SXM2 16 GB with 16 GB runs it at Q8_0, using about 11.7 GB and generating roughly 53.2 tokens per second — a comfortable fit.

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