Granite 3.2 2B TPS calculator

Open weights IBM 2.5B 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 · Q8_0 · 14.6 tok/s

Fastest card

B200

1,339 tok/s · 180 GB

Which GPUs can run Granite 3.2 2B?

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
1,339 tok/s 804–2,143 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	3.4 GB	Q8_0	Comfortable
1,339 tok/s 804–2,143 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	3.4 GB	Q8_0	Comfortable
1,069 tok/s 642–1,711 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	3.4 GB	Q8_0	Comfortable
1,069 tok/s 642–1,711 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	3.4 GB	Q8_0	Comfortable
855 tok/s 513–1,368 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	3.4 GB	Q8_0	Comfortable
819 tok/s 491–1,310 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	3.4 GB	Q8_0	Comfortable
819 tok/s 491–1,310 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	3.4 GB	Q8_0	Comfortable
783 tok/s 470–1,254 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	3.4 GB	Q8_0	Comfortable
695 tok/s 417–1,112 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	3.4 GB	Q8_0	Comfortable
695 tok/s 417–1,112 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	3.4 GB	Q8_0	Comfortable
695 tok/s 417–1,112 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	3.4 GB	Q8_0	Comfortable
660 tok/s 396–1,055 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	3.4 GB	Q8_0	Comfortable
562 tok/s 337–900 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	3.4 GB	Q8_0	Comfortable
562 tok/s 337–900 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	3.4 GB	Q8_0	Comfortable
562 tok/s 337–900 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	3.4 GB	Q8_0	Comfortable
562 tok/s 337–900 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	3.4 GB	Q8_0	Comfortable
562 tok/s 337–900 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	3.4 GB	Q8_0	Comfortable
428 tok/s 257–685 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	3.4 GB	Q8_0	Comfortable
428 tok/s 257–685 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	3.4 GB	Q8_0	Comfortable
357 tok/s 214–571 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	3.4 GB	Q8_0	Comfortable
349 tok/s 210–559 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	3.4 GB	Q8_0	Comfortable
342 tok/s 205–546 · low confidence	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	3.4 GB	Q8_0	Comfortable
342 tok/s 205–546 · low confidence	A100X NVIDIA	80 GB	2,040 GB/s	Jun 2021	3.4 GB	Q8_0	Comfortable
342 tok/s 205–546 · low confidence	A800 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Aug 2022	3.4 GB	Q8_0	Comfortable
342 tok/s 205–546 · low confidence	H100 CNX NVIDIA	80 GB	2,040 GB/s	Mar 2023	3.4 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: IBM
Organisation type: Industry
Country: United States of America
Published: 26 February 2025

What it does

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

Domain: Language
Task: Mathematical reasoning, Quantitative reasoning, Language modeling/generation, Question answering, Text summarization, Text classification, Translation, Code generation
Base model: Granite 3.1 2B

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: 2.5B
Training data: 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: NVIDIA H100 SXM5 80GB
Cloud vendor: IBM
Data centre: IBM's super computing cluster, Blue Vela, which is outfitted with NVIDIA H100 GPUs. This cluster provides a scalable and efficient infrastructure for training our models over thousands of GPUs.

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: ibm-granite

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: IBM Granite 3.2: Reasoning, vision, forecasting and more
Last updated: 28 November 2025

The extremes

The ten fastest GPUs that run Granite 3.2 2B

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 Granite 3.2 2B

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

Tesla C1080

Memory needed

3.4 GB

Fastest

1,339 tok/s

Granite 3.2 2B is small enough at 2.5B parameters that hardware is rarely the obstacle — 818 of the cards we track can run it, including cards several years old.

The entry point is the Tesla C1080: 4 GB of memory, Q8_0 compression, roughly 14.6 tokens per second.

The quickest result comes from a B200 at around 1,339 tokens per second — its 8,000 GB/s of bandwidth is what buys that.

What this model is

Granite 3.2 2B was published by IBM, in United States of America, in February 2025. It comes out of industry.

It works in Language, and is recorded as doing mathematical reasoning, Quantitative reasoning, Language modeling/generation, Question answering, Text summarization, Text classification, Translation, Code generation.

Its starting point was Granite 3.1 2B — most models at this scale are adapted from an existing base rather than built from nothing.

The weights being open is what puts this page in the calculator rather than only in the catalogue: it is a model you can actually hold. It is published under the ibm-granite organisation on Hugging Face.

What decides the speed

The median result is around 37.6 tokens per second; 783 cards produce text faster than most people read it.

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

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.

Step by step

How to choose a GPU for Granite 3.2 2B

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

01

Check what it needs before anything else

Every card here has been checked against Granite 3.2 2B — around 3.4 GB at Q8_0. Capacity is the gate — a card either holds it or it does not.
02

Match the context to your actual use

Set the context to what you will actually use. The cache grows with the conversation, and it is the usual reason Granite 3.2 2B stops fitting a card that seemed fine.
03

Decide how much compression you will accept

Compression is what makes Granite 3.2 2B fit smaller cards, at some cost in accuracy — Q8_0 on the smallest card that fits. A minimum quality removes the ones that go too far.
04

Compare tokens per second, not specifications

Sort by speed to see how cards rank for Granite 3.2 2B. It will not match a gaming ordering — generation is bound by memory bandwidth, which is why the B200 tops it at 1,339 tok/s.
05

Look at the headroom, not just the fit

The fit column separates cards that just manage Granite 3.2 2B from those with room to spare. Buy for the second if the context might grow.
06

Check the card from the other side

Every card name links to its own page, which runs the same calculation across the whole model catalogue. Worth a look before buying for Granite 3.2 2B alone — a card is usually bought for more than one model.

Answers

Granite 3.2 2B — common questions

Can I run Granite 3.2 2B if it does not fit in my GPU?

It can be split between the card and system memory, but Granite 3.2 2B generates painfully slowly that way. Nothing on this page assumes offloading.

Would two GPUs run Granite 3.2 2B faster?

Two cards buy memory rather than speed. That matters for Granite 3.2 2B only if one card cannot hold it — 818 can, so a second adds little.

Why does the quantisation differ between cards for Granite 3.2 2B?

Each card is shown running the least-compressed copy it can hold, and Granite 3.2 2B appears at 1 different compression levels across the cards that fit it. Bigger cards get the more accurate version.

How accurate are these Granite 3.2 2B speed estimates?

These are estimates with real error bars. The fastest result here, 804–2,143 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 Granite 3.2 2B?

The smallest card in our catalogue that holds Granite 3.2 2B is the Tesla C1080, with 4 GB of memory. It runs the model at Q8_0 using about 3.4 GB, and produces roughly 14.6 tokens per second. 818 cards in total can run it.

How fast is Granite 3.2 2B on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 1,339 tokens per second; the slowest that still runs it manages considerably less. Reading speed is around ten tokens per second, and 783 of the cards that can run Granite 3.2 2B clear that.

How much VRAM does Granite 3.2 2B need?

About 3.4 GB at Q8_0 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 Granite 3.2 2B on a 8 GB GPU?

Yes. A CMP 170HX 8 GB with 8 GB runs it at Q8_0, using about 3.4 GB and generating roughly 249 tokens per second — a comfortable fit.

Can I run Granite 3.2 2B on a 12 GB GPU?

Yes. A GeForce RTX 3080 Ti with 12 GB runs it at Q8_0, using about 3.4 GB and generating roughly 153 tokens per second — a comfortable fit.

Can I run Granite 3.2 2B on a 16 GB GPU?

Yes. A Tesla V100 SXM2 16 GB with 16 GB runs it at Q8_0, using about 3.4 GB and generating roughly 189 tokens per second — a comfortable fit.

Can I run Granite 3.2 2B on a 24 GB GPU?

Yes. A GeForce RTX 5090 D V2 with 24 GB runs it at Q8_0, using about 3.4 GB and generating roughly 224 tokens per second — a comfortable fit.

Is Granite 3.2 2B open source?

Its weights are published, so Granite 3.2 2B 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 Granite 3.2 2B have?

Granite 3.2 2B has 2.5B 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 Granite 3.2 2B?

Granite 3.2 2B was published by IBM, based in United States of America, categorised as industry.

When was Granite 3.2 2B released?

Granite 3.2 2B was published in February 2025.

What is Granite 3.2 2B used for?

Granite 3.2 2B works in Language, and is recorded as handling mathematical reasoning, Quantitative reasoning, Language modeling/generation, Question answering, Text summarization, Text classification, Translation, Code 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 Granite 3.2 2B?

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

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