Mistral Small 3.2 TPS calculator

Open weights Mistral AI 24B parameters June 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

241 cards that can run it

818 cards we hold specifications for

Smallest card that fits

Xeon Phi 7120P

16 GB · IQ4_XS · 9.9 tok/s

Fastest card

B200

141 tok/s · 180 GB

Which GPUs can run Mistral Small 3.2?

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

241 cards match

Calculating

					Needs	Quantisation	Fit
141 tok/s 120–169	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	26.1 GB	Q8_0	Comfortable
141 tok/s 120–169	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	26.1 GB	Q8_0	Comfortable
113 tok/s 68–180 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	26.1 GB	Q8_0	Comfortable
113 tok/s 68–180 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	26.1 GB	Q8_0	Comfortable
90.2 tok/s 54–144 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	26.1 GB	Q8_0	Comfortable
86.3 tok/s 73–104	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	26.1 GB	Q8_0	Comfortable
86.3 tok/s 73–104	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	26.1 GB	Q8_0	Comfortable
82.6 tok/s 50–132 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	26.1 GB	Q8_0	Comfortable
73.3 tok/s 44–117 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	26.1 GB	Q8_0	Comfortable
73.3 tok/s 44–117 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	26.1 GB	Q8_0	Comfortable
73.3 tok/s 44–117 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	26.1 GB	Q8_0	Comfortable
69.5 tok/s 59–83	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	26.1 GB	Q8_0	Comfortable
59.3 tok/s 50–71	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	26.1 GB	Q8_0	Comfortable
59.3 tok/s 50–71	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	26.1 GB	Q8_0	Comfortable
59.3 tok/s 50–71	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	26.1 GB	Q8_0	Comfortable
59.3 tok/s 50–71	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	26.1 GB	Q8_0	Comfortable
59.3 tok/s 50–71	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	26.1 GB	Q8_0	Comfortable
49.0 tok/s 42–59	Tesla V100 SXM2 16 GB NVIDIA	16 GB	1,130 GB/s	Nov 2019	13.6 GB	IQ4_XS	Tight
45.2 tok/s 27–72 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	26.1 GB	Q8_0	Comfortable
45.2 tok/s 27–72 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	26.1 GB	Q8_0	Comfortable
41.6 tok/s 35–50	GeForce RTX 5080 NVIDIA	16 GB	960 GB/s	Jan 2025	13.6 GB	IQ4_XS	Tight
38.9 tok/s 33–47	Tesla V100 DGXS 16 GB NVIDIA	16 GB	897 GB/s	Mar 2018	13.6 GB	IQ4_XS	Tight
38.9 tok/s 33–47	Tesla V100 PCIe 16 GB NVIDIA	16 GB	897 GB/s	Jun 2017	13.6 GB	IQ4_XS	Tight
38.9 tok/s 33–47	Tesla V100 SXM2 16 GB NVIDIA	16 GB	897 GB/s	Jun 2017	13.6 GB	IQ4_XS	Tight
38.8 tok/s 33–47	GeForce RTX 5070 Ti NVIDIA	16 GB	896 GB/s	Feb 2025	13.6 GB	IQ4_XS	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: Mistral AI
Organisation type: Industry
Country: France
Published: 15 June 2025

What it does

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

Domain: Language, Vision, Multimodal
Task: Language modeling/generation, Question answering, Quantitative reasoning, Code generation, Translation, Visual question answering, Character recognition (OCR)
Base model: Mistral Small 3.1

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: 24B
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: mistralai

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: Mistral-Small-3.2-24B-Instruct-2506 is a minor update of Mistral-Small-3.1-24B-Instruct-2503.
Last updated: 28 November 2025

The extremes

The ten fastest GPUs that run Mistral Small 3.2

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 Mistral Small 3.2

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 7120P

Memory needed

13.6 GB

Fastest

141 tok/s

Mistral Small 3.2 reaches a parameter count of 24B. That lands in the range a serious desktop card can handle once the weights are compressed. The number of cards we track that can run it: 241.

The entry point is Xeon Phi 7120P, with a memory capacity of 16 GB, running it at a compression of IQ4_XS and producing around 9.9 tokens per second.

At the other end sits B200, generating roughly 141 tokens per second on the strength of a memory bandwidth of 8,000 GB/s.

Where it came from

Mistral Small 3.2 was published by Mistral AI, in the country recorded as France, during June 2025. The publishing organisation is categorised as industry.

It works in the domain of Language, Vision, Multimodal, and is recorded as performing the task of language modeling/generation, Question answering, Quantitative reasoning, Code generation, Translation, Visual question answering, Character recognition (OCR).

Its starting point was an existing base model, Mistral Small 3.1. 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 mistralai.

Understanding the speeds

Across every card that can run it, the middle of the range sits at 19.4 tokens per second. Clearing the ten tokens per second that roughly matches reading speed: 192 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.

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

Step by step

How to choose a GPU for Mistral Small 3.2

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 Mistral Small 3.2, needing around 13.6 GB at a compression of IQ4_XS. No amount of processing power compensates for a card that cannot hold it.
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 Mistral Small 3.2.
03

Set a quality floor

The quantisation column varies by card, because a bigger card holds a more accurate copy, reaching a compression of IQ4_XS 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

Ranking by tokens per second follows memory bandwidth rather than core counts, for Mistral Small 3.2. It will not match a gaming ordering, because generation is bound by memory bandwidth. The card topping the list is B200, at 141 tok/s.
05

Look at the headroom, not just the fit

The fit column separates cards that just manage it from those with room to spare, in the case of Mistral Small 3.2. 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

Check the card from the other side

Each card page repeats this sweep for every model we hold, answering what else the hardware is good for beyond Mistral Small 3.2.

Answers

Mistral Small 3.2 — common questions

Mistral Small 3.2— who created it?

It was published by Mistral AI, based in France, an organisation categorised as industry.

Mistral Small 3.2— when was it released?

It was published in June 2025.

Mistral Small 3.2— what is it used for?

It works in the domain of Language, Vision, Multimodal, and is recorded as handling the task of language modeling/generation, Question answering, Quantitative reasoning, Code generation, Translation, Visual question answering, Character recognition (OCR). Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.

Mistral Small 3.2— where can I download it?

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

Mistral Small 3.2— 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. The nearest miss we calculate falls short by 4.2 GB. Every figure here assumes the whole model is resident on the card.

Mistral Small 3.2— 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: 241. So a second card is rarely the answer here.

Mistral Small 3.2— 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: 4. Bigger cards get the more accurate version, and the quality floor above pins the comparison to one level.

Mistral Small 3.2— how accurate are these speed estimates?

These are estimates with real error bars, and any of them could reasonably land anywhere in its published range depending on which runtime you use. The fastest result here: 120–169 tok/s on B200. The same model and card vary by thirty to fifty per cent depending on the inference software and its version.

Mistral Small 3.2— what GPU do I need to run it?

The smallest card in our catalogue that holds it is Xeon Phi 7120P, with a memory capacity of 16 GB. It runs the model at a compression of IQ4_XS using about 13.6 GB, and produces roughly 9.9 tokens per second. The number of cards able to run it in total: 241.

Mistral Small 3.2— how fast is it on a GPU?

It depends on the card. The quickest we calculate is B200, at about 141 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: 192.

Mistral Small 3.2— how much VRAM does it need?

It needs about 13.6 GB at a compression of IQ4_XS, 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.

Mistral Small 3.2— 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 IQ4_XS, using about 13.6 GB and generating roughly 49.0 tokens per second. The fit is tight.

Mistral Small 3.2— 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 Q6_K, using about 20.6 GB and generating roughly 34.4 tokens per second. The fit is tight.

Mistral Small 3.2— 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.

Mistral Small 3.2— how many parameters does it have?

It has a parameter count of 24B. 24B. 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.

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