Qwen3-8B TPS calculator

Open weights Alibaba 8.2B parameters April 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

582 of 818 cards that can run it

Smallest card that fits

Quadro 6000

6 GB · Q3_K_M · 17.0 tok/s

Fastest card

B200

413 tok/s · 180 GB

Which GPUs can run Qwen3-8B?

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

582 cards match

Calculating

					Needs	Quantisation	Fit
413 tok/s 351–496	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	10.1 GB	Q8_0	Comfortable
413 tok/s 351–496	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	10.1 GB	Q8_0	Comfortable
330 tok/s 198–528 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	10.1 GB	Q8_0	Comfortable
330 tok/s 198–528 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	10.1 GB	Q8_0	Comfortable
264 tok/s 158–422 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	10.1 GB	Q8_0	Comfortable
253 tok/s 215–303	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	10.1 GB	Q8_0	Comfortable
253 tok/s 215–303	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	10.1 GB	Q8_0	Comfortable
242 tok/s 145–387 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	10.1 GB	Q8_0	Comfortable
215 tok/s 129–343 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	10.1 GB	Q8_0	Comfortable
215 tok/s 129–343 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	10.1 GB	Q8_0	Comfortable
215 tok/s 129–343 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	10.1 GB	Q8_0	Comfortable
204 tok/s 173–244	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	10.1 GB	Q8_0	Comfortable
178 tok/s 151–213	CMP 170HX 8 GB NVIDIA	8 GB	1,490 GB/s	Sep 2021	6.3 GB	Q4_K_M	Tight
174 tok/s 148–208	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	10.1 GB	Q8_0	Comfortable
174 tok/s 148–208	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	10.1 GB	Q8_0	Comfortable
174 tok/s 148–208	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	10.1 GB	Q8_0	Comfortable
174 tok/s 148–208	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	10.1 GB	Q8_0	Comfortable
174 tok/s 148–208	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	10.1 GB	Q8_0	Comfortable
132 tok/s 79–211 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	10.1 GB	Q8_0	Comfortable
132 tok/s 79–211 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	10.1 GB	Q8_0	Comfortable
117 tok/s 100–140	CMP 170HX 10 GB NVIDIA	10 GB	1,560 GB/s	Sep 2021	8.2 GB	Q6_K	Tight
110 tok/s 66–176 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	10.1 GB	Q8_0	Comfortable
108 tok/s 65–172 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	10.1 GB	Q8_0	Comfortable
105 tok/s 90–126	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	10.1 GB	Q8_0	Comfortable
105 tok/s 90–126	A100X NVIDIA	80 GB	2,040 GB/s	Jun 2021	10.1 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: 29 April 2025

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, Mathematical reasoning, Quantitative reasoning, Code generation, Translation

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: 8.2B
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: 1.8 × 10²⁴ 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 (unrestricted)
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: Qwen3: Think Deeper, Act Faster
Last updated: 28 November 2025

The extremes

The ten fastest GPUs for Qwen3-8B

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 Qwen3-8B

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

The hardware side

Minimum card

Quadro 6000

Memory needed

5.3 GB

Fastest

413 tok/s

Qwen3-8B is small enough at 8.2B parameters that hardware is rarely the obstacle — 582 of the cards we track can run it, including cards several years old.

At the low end, a Quadro 6000 handles it — 6 GB, at Q3_K_M, for about 17.0 tokens per second.

A B200 is the fastest we calculate for it: about 413 tokens per second, from 8,000 GB/s of memory bandwidth.

About this model

Qwen3-8B was published by Alibaba, in China, in April 2025. It comes out of industry.

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

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

The median result is around 26.0 tokens per second; 554 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.

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

How it was trained

Producing it required around 1.8 × 10²⁴ FLOP of arithmetic, which is a statement about the training budget rather than about inference.

Step by step

How to choose a GPU for Qwen3-8B

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 Qwen3-8B — around 5.3 GB at Q3_K_M. That figure, not the card's headline performance, is what decides whether it runs.
02

Match the context to your actual use

Longer conversations cost memory on top of what the weights need. Move the slider to your real working length before trusting any row for Qwen3-8B.
03

Set a quality floor

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 Qwen3-8B by squeezing it further than you would want.
04

Compare tokens per second, not specifications

Ranking by tokens per second for Qwen3-8B follows memory bandwidth, not core counts, which is why the B200 tops it at 413 tok/s.
05

Look at the headroom, not just the fit

A tight fit runs Qwen3-8B 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

Each card page repeats this sweep for every model we hold. It answers what else the hardware is good for, beyond Qwen3-8B.

Answers

Qwen3-8B — common questions

Would two GPUs run Qwen3-8B faster?

Two cards buy memory rather than speed. That matters for Qwen3-8B only if one card cannot hold it — 582 can, so a second adds little.

Why does the quantisation differ between cards for Qwen3-8B?

Each card is shown running the least-compressed copy it can hold, and Qwen3-8B appears at 4 different compression levels across the cards that fit it. Bigger cards get the more accurate version.

How accurate are these Qwen3-8B speed estimates?

These are estimates with real error bars. The fastest result here, 351–496 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 Qwen3-8B?

The smallest card in our catalogue that holds Qwen3-8B is the Quadro 6000, with 6 GB of memory. It runs the model at Q3_K_M using about 5.3 GB, and produces roughly 17.0 tokens per second. 582 cards in total can run it.

How fast is Qwen3-8B on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 413 tokens per second; the slowest that still runs it manages considerably less. Reading speed is around ten tokens per second, and 554 of the cards that can run Qwen3-8B clear that.

How much VRAM does Qwen3-8B need?

About 5.3 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 Qwen3-8B on a 8 GB GPU?

Yes. A CMP 170HX 8 GB with 8 GB runs it at Q4_K_M, using about 6.3 GB and generating roughly 178 tokens per second — a tight fit.

Can I run Qwen3-8B on a 12 GB GPU?

Yes. A GeForce RTX 3080 Ti with 12 GB runs it at Q8_0, using about 10.1 GB and generating roughly 47.1 tokens per second — a tight fit.

Can I run Qwen3-8B on a 16 GB GPU?

Yes. A Tesla V100 SXM2 16 GB with 16 GB runs it at Q8_0, using about 10.1 GB and generating roughly 58.4 tokens per second — a comfortable fit.

Can I run Qwen3-8B on a 24 GB GPU?

Yes. A GeForce RTX 5090 D V2 with 24 GB runs it at Q8_0, using about 10.1 GB and generating roughly 69.2 tokens per second — a comfortable fit.

Is Qwen3-8B open source?

Its weights are published, so Qwen3-8B 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 Qwen3-8B have?

Qwen3-8B has 8.2B parameters. Number of Parameters: 8.2B Number of Paramaters (Non-Embedding): 6.95B Number of Layers: 36 Number of Attention Heads (GQA): 32 for Q and 8 for KV Context Length: 32,768 natively and 131,072 tokens with YaRN. 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 Qwen3-8B?

Qwen3-8B was published by Alibaba, based in China, categorised as industry.

When was Qwen3-8B released?

Qwen3-8B was published in April 2025.

What is Qwen3-8B used for?

Qwen3-8B works in Language, and is recorded as handling language modeling/generation, Question answering, Mathematical reasoning, Quantitative reasoning, Code generation, Translation. These are the areas it was designed around; they describe intent rather than a hard boundary.

Where can I download Qwen3-8B?

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 Qwen3-8B?

Around 1.8 × 10²⁴ 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 Qwen3-8B 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 1.8 GB. Our figures for Qwen3-8B assume it is fully resident.

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