Qwen-Audio-Chat TPS calculator

Open weights Alibaba 8.5B parameters November 2023

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 cards that can run it

818 cards we hold specifications for

Smallest card that fits

Quadro 6000

6 GB · IQ4_XS · 15.0 tok/s

Fastest card

B200

401 tok/s · 180 GB

Which GPUs can run Qwen-Audio-Chat?

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
401 tok/s 240–641 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	9.8 GB	Q8_0	Comfortable
401 tok/s 240–641 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	9.8 GB	Q8_0	Comfortable
320 tok/s 192–512 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	9.8 GB	Q8_0	Comfortable
320 tok/s 192–512 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	9.8 GB	Q8_0	Comfortable
256 tok/s 153–409 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	9.8 GB	Q8_0	Comfortable
245 tok/s 147–392 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	9.8 GB	Q8_0	Comfortable
245 tok/s 147–392 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	9.8 GB	Q8_0	Comfortable
234 tok/s 141–375 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	9.8 GB	Q8_0	Comfortable
208 tok/s 125–333 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	9.8 GB	Q8_0	Comfortable
208 tok/s 125–333 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	9.8 GB	Q8_0	Comfortable
208 tok/s 125–333 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	9.8 GB	Q8_0	Comfortable
197 tok/s 118–316 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	9.8 GB	Q8_0	Comfortable
168 tok/s 101–269 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	9.8 GB	Q8_0	Comfortable
168 tok/s 101–269 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	9.8 GB	Q8_0	Comfortable
168 tok/s 101–269 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	9.8 GB	Q8_0	Comfortable
168 tok/s 101–269 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	9.8 GB	Q8_0	Comfortable
168 tok/s 101–269 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	9.8 GB	Q8_0	Comfortable
133 tok/s 80–213 · low confidence	CMP 170HX 8 GB NVIDIA	8 GB	1,490 GB/s	Sep 2021	6.8 GB	Q5_K_M	Tight
128 tok/s 77–205 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	9.8 GB	Q8_0	Comfortable
128 tok/s 77–205 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	9.8 GB	Q8_0	Comfortable
113 tok/s 68–182 · low confidence	CMP 170HX 10 GB NVIDIA	10 GB	1,560 GB/s	Sep 2021	7.8 GB	Q6_K	Tight
107 tok/s 64–171 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	9.8 GB	Q8_0	Comfortable
104 tok/s 63–167 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	9.8 GB	Q8_0	Comfortable
102 tok/s 61–163 · low confidence	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	9.8 GB	Q8_0	Comfortable
102 tok/s 61–163 · low confidence	A100X NVIDIA	80 GB	2,040 GB/s	Jun 2021	9.8 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: 14 November 2023
Authors: Yunfei Chu, Jin Xu, Xiaohuan Zhou, Qian Yang, Shiliang Zhang, Zhijie Yan, Chang Zhou, Jingren Zhou

What it does

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

Domain: Language, Speech, Audio
Task: Audio question answering, Chat, Speech recognition (ASR), Translation, Transcription, Text classification, Question answering, Audio classification, Voice identification, Part-of-speech tagging, Speech-to-speech
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: 8.5B
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 (restricted use)
Training code: Open (restricted use)

How it is classified

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

Why it is tracked: SOTA improvement
Record confidence: Likely
Citations: 743

Sources

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

Reference: Qwen-Audio: Advancing Universal Audio Understanding via Unified Large-Scale Audio-Language Models
Last updated: 25 May 2026

The extremes

The ten fastest GPUs that run Qwen-Audio-Chat

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 Qwen-Audio-Chat

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

401 tok/s

Qwen-Audio-Chat is small enough at 8.5B 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 IQ4_XS, for about 15.0 tokens per second.

The quickest result comes from a B200 at around 401 tokens per second — its 8,000 GB/s of bandwidth is what buys that.

Background

Qwen-Audio-Chat was published by Alibaba, in China, in November 2023. industry is the category the publisher falls under.

It works in Language, Speech, Audio, and is recorded as doing audio question answering, Chat, Speech recognition (ASR), Translation, Transcription, Text classification, Question answering, Audio classification, Voice identification, Part-of-speech tagging, Speech-to-speech.

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.

Reading the throughput figures

The median result is around 22.5 tokens per second; 549 cards produce text faster than most people read it.

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.

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.

Training and provenance

The reason it appears in this catalogue at all is sOTA improvement.

Step by step

How to choose a GPU for Qwen-Audio-Chat

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

Every card here has been checked against Qwen-Audio-Chat — around 5.3 GB at IQ4_XS. Capacity is the gate — a card either holds it or it does not.
02

Set the context length you will work at

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

Choose how far you will compress it

The quantisation column varies by card, because a bigger card holds a more accurate copy of Qwen-Audio-Chat — IQ4_XS on the smallest card that fits. Set a floor to hold the comparison at one level.
04

Compare tokens per second, not specifications

Ranking by tokens per second for Qwen-Audio-Chat follows memory bandwidth, not core counts, which is why the B200 tops it at 401 tok/s.
05

Read the fit column last

A tight fit runs Qwen-Audio-Chat 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

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

Answers

Qwen-Audio-Chat — common questions

Is Qwen-Audio-Chat open source?

Its weights are published, so Qwen-Audio-Chat 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 Qwen-Audio-Chat have?

Qwen-Audio-Chat has 8.5B parameters. the model has two components - audio and language. 670M + 7.7B = 8.46B "The audio encoder is composed of 640M parameters" "Qwen-Audio incorporates a large language model as its foundational component. The model is initialized using pre-trained weights derived from Qwen-7B (Bai et al., 2023a). Qwen-7B is a 32-layer Transformer decoder model with a hidden size of 4096, encompassing a total of 7.7B 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 Qwen-Audio-Chat?

Qwen-Audio-Chat was published by Alibaba, based in China, categorised as industry.

When was Qwen-Audio-Chat released?

Qwen-Audio-Chat was published in November 2023. 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 Qwen-Audio-Chat used for?

Qwen-Audio-Chat works in Language, Speech, Audio, and is recorded as handling audio question answering, Chat, Speech recognition (ASR), Translation, Transcription, Text classification, Question answering, Audio classification, Voice identification, Part-of-speech tagging, Speech-to-speech. Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.

Where can I download Qwen-Audio-Chat?

The weights for Qwen-Audio-Chat are published, though we do not hold a repository link for it. This site calculates hardware requirements rather than hosting model files.

Can I run Qwen-Audio-Chat if it does not fit in my GPU?

Partly. Layers that do not fit sit in system memory and run at a fraction of the speed, so a mostly-offloaded Qwen-Audio-Chat is rarely worth using — the nearest miss we calculate is short by 1.3 GB. Every figure here assumes the whole model is on the card.

Would two GPUs run Qwen-Audio-Chat faster?

A second card roughly doubles the memory available but not the generation rate. With 582 cards already able to run Qwen-Audio-Chat alone, the case for pairing is weak.

Why does the quantisation differ between cards for Qwen-Audio-Chat?

Because capacity varies, so does how hard Qwen-Audio-Chat has to be squeezed — 4 distinct levels appear in the table above. Set a minimum quality to compare at one.

How accurate are these Qwen-Audio-Chat speed estimates?

They are calculated from specifications rather than measured, and each carries a range — 240–641 tok/s on the B200, for instance. The same model and card vary by thirty to fifty per cent depending on the inference software and its version.

What GPU do I need to run Qwen-Audio-Chat?

The smallest card in our catalogue that holds Qwen-Audio-Chat is the Quadro 6000, with 6 GB of memory. It runs the model at IQ4_XS using about 5.3 GB, and produces roughly 15.0 tokens per second. 582 cards in total can run it.

How fast is Qwen-Audio-Chat on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 401 tokens per second; the slowest that still runs it manages considerably less. Reading speed is around ten tokens per second, and 549 of the cards that can run Qwen-Audio-Chat clear that.

How much VRAM does Qwen-Audio-Chat need?

About 5.3 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 Qwen-Audio-Chat on a 8 GB GPU?

Yes. A CMP 170HX 8 GB with 8 GB runs it at Q5_K_M, using about 6.8 GB and generating roughly 133 tokens per second — a tight fit.

Can I run Qwen-Audio-Chat on a 12 GB GPU?

Yes. A GeForce RTX 3080 Ti with 12 GB runs it at Q8_0, using about 9.8 GB and generating roughly 45.7 tokens per second — a tight fit.

Can I run Qwen-Audio-Chat on a 16 GB GPU?

Yes. A Tesla V100 SXM2 16 GB with 16 GB runs it at Q8_0, using about 9.8 GB and generating roughly 56.6 tokens per second — a comfortable fit.

Can I run Qwen-Audio-Chat on a 24 GB GPU?

Yes. A GeForce RTX 5090 D V2 with 24 GB runs it at Q8_0, using about 9.8 GB and generating roughly 67.1 tokens per second — a comfortable fit.

Source

Original publication

Record last updated 25 May 2026

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