Qwen2.5-3B TPS calculator

Open weights Alibaba 3.1B parameters September 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

818 cards that can run it

818 cards we hold specifications for

Smallest card that fits

Tesla C1080

4 GB · Q6_K · 17.3 tok/s

Fastest card

B200

1,097 tok/s · 180 GB

Which GPUs can run Qwen2.5-3B?

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
1,097 tok/s 932–1,316	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	4.1 GB	Q8_0	Comfortable
1,097 tok/s 932–1,316	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	4.1 GB	Q8_0	Comfortable
876 tok/s 525–1,401 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	4.1 GB	Q8_0	Comfortable
876 tok/s 525–1,401 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	4.1 GB	Q8_0	Comfortable
700 tok/s 420–1,120 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	4.1 GB	Q8_0	Comfortable
670 tok/s 570–804	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	4.1 GB	Q8_0	Comfortable
670 tok/s 570–804	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	4.1 GB	Q8_0	Comfortable
641 tok/s 385–1,026 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	4.1 GB	Q8_0	Comfortable
569 tok/s 342–911 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	4.1 GB	Q8_0	Comfortable
569 tok/s 342–911 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	4.1 GB	Q8_0	Comfortable
569 tok/s 342–911 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	4.1 GB	Q8_0	Comfortable
540 tok/s 459–648	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	4.1 GB	Q8_0	Comfortable
461 tok/s 391–553	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	4.1 GB	Q8_0	Comfortable
461 tok/s 391–553	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	4.1 GB	Q8_0	Comfortable
461 tok/s 391–553	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	4.1 GB	Q8_0	Comfortable
461 tok/s 391–553	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	4.1 GB	Q8_0	Comfortable
461 tok/s 391–553	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	4.1 GB	Q8_0	Comfortable
351 tok/s 210–561 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	4.1 GB	Q8_0	Comfortable
351 tok/s 210–561 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	4.1 GB	Q8_0	Comfortable
292 tok/s 175–468 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	4.1 GB	Q8_0	Comfortable
286 tok/s 172–458 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	4.1 GB	Q8_0	Comfortable
280 tok/s 238–336	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	4.1 GB	Q8_0	Comfortable
280 tok/s 238–336	A100X NVIDIA	80 GB	2,040 GB/s	Jun 2021	4.1 GB	Q8_0	Comfortable
280 tok/s 238–336	A800 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Aug 2022	4.1 GB	Q8_0	Comfortable
280 tok/s 238–336	H100 CNX NVIDIA	80 GB	2,040 GB/s	Mar 2023	4.1 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: Alibaba
Organisation type: Industry
Country: China
Published: 19 September 2024
Authors: Qwen Team

What it does

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

Domain: Language
Task: Language modeling/generation, Question answering, Quantitative reasoning

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: 3.1B
Training data: tokens
Epochs: 1

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: 3.3 × 10²³ FLOP
How it was established: Operation counting

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 (non-commercial)
Training code: Unreleased
Hugging Face: Qwen

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: Qwen2.5: A Party of Foundation Models!
Last updated: 28 November 2025

The extremes

The ten fastest GPUs that run Qwen2.5-3B

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 Qwen2.5-3B

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

What you need to run it

Minimum card

Tesla C1080

Memory needed

3.4 GB

Fastest

1,097 tok/s

Qwen2.5-3B is small enough at 3.1B parameters that hardware is rarely the obstacle — 818 of the cards we track can run it, including cards several years old.

The least hardware that works is a Tesla C1080. Its 4 GB is enough at Q6_K compression, giving roughly 17.3 tokens per second.

At the other end, a B200 generates roughly 1,097 tokens per second on it, on the strength of 8,000 GB/s of memory bandwidth.

About this model

Qwen2.5-3B was published by Alibaba, in China, in September 2024. industry is the category the publisher falls under.

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

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 Qwen organisation on Hugging Face.

How fast it runs, and why

Across every card that can run it, the middle of the range is about 34.8 tokens per second, and 779 of them clear the ten tokens per second that roughly matches reading speed.

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.

Because the architecture is recorded, the memory column is derived rather than estimated.

Training and provenance

Producing it required around 3.3 × 10²³ FLOP of arithmetic, which is a statement about the training budget rather than about inference.

Step by step

How to choose a GPU for Qwen2.5-3B

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

Look at what Qwen2.5-3B actually needs — around 3.4 GB at Q6_K. No amount of processing power compensates for a card that cannot hold it.
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 Qwen2.5-3B can slip off a card that handles short questions easily.
03

Decide how much compression you will accept

Each card runs the least-compressed copy it can hold — Q6_K on the smallest card that fits. Setting a floor drops the cards that only manage Qwen2.5-3B by squeezing it further than you would want.
04

Rank by throughput rather than spec sheet

Ranking by tokens per second for Qwen2.5-3B follows memory bandwidth, not core counts, which is why the B200 tops it at 1,097 tok/s.
05

Read the fit column last

Tight means Qwen2.5-3B 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

See what else that card runs

Each card page repeats this sweep for every model we hold. It answers what else the hardware is good for, beyond Qwen2.5-3B.

Answers

Qwen2.5-3B — common questions

Can I run Qwen2.5-3B on a 16 GB GPU?

Yes. A Tesla V100 SXM2 16 GB with 16 GB runs it at Q8_0, using about 4.1 GB and generating roughly 155 tokens per second — a comfortable fit.

Can I run Qwen2.5-3B on a 24 GB GPU?

Yes. A GeForce RTX 5090 D V2 with 24 GB runs it at Q8_0, using about 4.1 GB and generating roughly 184 tokens per second — a comfortable fit.

Is Qwen2.5-3B open source?

Its weights are published, so Qwen2.5-3B 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 Qwen2.5-3B have?

Qwen2.5-3B has 3.1B parameters. 3.09B. 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 Qwen2.5-3B?

Qwen2.5-3B was published by Alibaba, based in China, categorised as industry.

When was Qwen2.5-3B released?

Qwen2.5-3B was published in September 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 Qwen2.5-3B used for?

Qwen2.5-3B works in Language, and is recorded as handling language modeling/generation, Question answering, Quantitative reasoning. These are the areas it was designed around; they describe intent rather than a hard boundary.

Where can I download Qwen2.5-3B?

Its weights are published under the Qwen 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 Qwen2.5-3B?

Around 3.3 × 10²³ FLOP. 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 Qwen2.5-3B if it does not fit in my GPU?

It can be split between the card and system memory, but Qwen2.5-3B generates painfully slowly that way. Nothing on this page assumes offloading.

Would two GPUs run Qwen2.5-3B faster?

Two cards buy memory rather than speed. That matters for Qwen2.5-3B only if one card cannot hold it — 818 can, so a second adds little.

Why does the quantisation differ between cards for Qwen2.5-3B?

A larger card holds a more accurate copy. Across the cards that run Qwen2.5-3B, 2 compression levels are used; the floor control above pins it to one.

How accurate are these Qwen2.5-3B 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 932–1,316 tok/s on the B200 rather than a single number.

What GPU do I need to run Qwen2.5-3B?

The smallest card in our catalogue that holds Qwen2.5-3B is the Tesla C1080, with 4 GB of memory. It runs the model at Q6_K using about 3.4 GB, and produces roughly 17.3 tokens per second. 818 cards in total can run it.

How fast is Qwen2.5-3B on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 1,097 tokens per second; the slowest that still runs it manages considerably less. Reading speed is around ten tokens per second, and 779 of the cards that can run Qwen2.5-3B clear that.

How much VRAM does Qwen2.5-3B need?

About 3.4 GB at Q6_K 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 Qwen2.5-3B on a 8 GB GPU?

Yes. A CMP 170HX 8 GB with 8 GB runs it at Q8_0, using about 4.1 GB and generating roughly 204 tokens per second — a comfortable fit.

Can I run Qwen2.5-3B on a 12 GB GPU?

Yes. A GeForce RTX 3080 Ti with 12 GB runs it at Q8_0, using about 4.1 GB and generating roughly 125 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