Qwen2-VL-2B TPS calculator

Open weights Alibaba 2B 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 · Q8_0 · 18.4 tok/s

Fastest card

B200

1,694 tok/s · 180 GB

Which GPUs can run Qwen2-VL-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
1,694 tok/s 1,016–2,711 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	2.8 GB	Q8_0	Comfortable
1,694 tok/s 1,016–2,711 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	2.8 GB	Q8_0	Comfortable
1,353 tok/s 812–2,164 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	2.8 GB	Q8_0	Comfortable
1,353 tok/s 812–2,164 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	2.8 GB	Q8_0	Comfortable
1,082 tok/s 649–1,731 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	2.8 GB	Q8_0	Comfortable
1,036 tok/s 621–1,657 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	2.8 GB	Q8_0	Comfortable
1,036 tok/s 621–1,657 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	2.8 GB	Q8_0	Comfortable
991 tok/s 595–1,586 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	2.8 GB	Q8_0	Comfortable
880 tok/s 528–1,407 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	2.8 GB	Q8_0	Comfortable
880 tok/s 528–1,407 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	2.8 GB	Q8_0	Comfortable
880 tok/s 528–1,407 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	2.8 GB	Q8_0	Comfortable
834 tok/s 501–1,335 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	2.8 GB	Q8_0	Comfortable
712 tok/s 427–1,138 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	2.8 GB	Q8_0	Comfortable
712 tok/s 427–1,138 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	2.8 GB	Q8_0	Comfortable
712 tok/s 427–1,138 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	2.8 GB	Q8_0	Comfortable
712 tok/s 427–1,138 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	2.8 GB	Q8_0	Comfortable
712 tok/s 427–1,138 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	2.8 GB	Q8_0	Comfortable
542 tok/s 325–867 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	2.8 GB	Q8_0	Comfortable
542 tok/s 325–867 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	2.8 GB	Q8_0	Comfortable
451 tok/s 271–722 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	2.8 GB	Q8_0	Comfortable
442 tok/s 265–707 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	2.8 GB	Q8_0	Comfortable
432 tok/s 259–691 · low confidence	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	2.8 GB	Q8_0	Comfortable
432 tok/s 259–691 · low confidence	A100X NVIDIA	80 GB	2,040 GB/s	Jun 2021	2.8 GB	Q8_0	Comfortable
432 tok/s 259–691 · low confidence	A800 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Aug 2022	2.8 GB	Q8_0	Comfortable
432 tok/s 259–691 · low confidence	H100 CNX NVIDIA	80 GB	2,040 GB/s	Mar 2023	2.8 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: 18 September 2024
Authors: Peng Wang, Shuai Bai, Sinan Tan, Shijie Wang, Zhihao Fan, Jinze Bai, Keqin Chen, Xuejing Liu, Jialin Wang, Wenbin Ge, Yang Fan, Kai Dang, Mengfei Du, Xuancheng Ren, Rui Men, Dayiheng Liu, Chang Zhou, Jingren Zhou, Junyang Lin

What it does

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

Domain: Language, Vision, Multimodal
Task: Visual question answering, Video description, Language modeling/generation, Translation, Question answering, Character recognition (OCR), 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: 2B
Training data: 1,400,000,000,000 tokens

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: 1.7 × 10²² 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.

Record confidence: Confident

Sources

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

Reference: Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution
Last updated: 28 November 2025

The extremes

The ten fastest GPUs that run Qwen2-VL-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 Qwen2-VL-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

2.8 GB

Fastest

1,694 tok/s

Qwen2-VL-2B is small enough at 2B 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 18.4 tokens per second.

Top of the range is the B200, at roughly 1,694 tokens per second thanks to 8,000 GB/s of bandwidth.

What this model is

Qwen2-VL-2B was published by Alibaba, in China, in September 2024. It comes out of industry.

It works in Language, Vision, Multimodal, and is recorded as doing visual question answering, Video description, Language modeling/generation, Translation, Question answering, Character recognition (OCR), Quantitative reasoning.

The weights are published, so it can be downloaded and run on your own hardware indefinitely, offline, with no account attached. It is published under the Qwen organisation on Hugging Face.

What decides the speed

The median result is around 47.6 tokens per second; 789 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.

Without the attention layout on record, the memory column is an approximation. It is close enough to choose hardware by, and least reliable at long context.

What went into building it

Training it took roughly 1.7 × 10²² FLOP of computation — a measure of what producing the model cost, not of how fast it answers.

The training set ran to roughly 1,400,000,000,000 tokens.

Step by step

How to choose a GPU for Qwen2-VL-2B

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

01

Read the memory figure first

Every card here has been checked against Qwen2-VL-2B — around 2.8 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

Longer conversations cost memory on top of what the weights need. Move the slider to your real working length before trusting any row for Qwen2-VL-2B.
03

Choose how far you will compress it

Each card runs the least-compressed copy it can hold — Q8_0 on the smallest card that fits. Setting a floor drops the cards that only manage Qwen2-VL-2B by squeezing it further than you would want.
04

Sort by speed

Sort by speed to see how cards rank for Qwen2-VL-2B. It will not match a gaming ordering — generation is bound by memory bandwidth, which is why the B200 tops it at 1,694 tok/s.
05

Check the fit verdict before buying

A tight fit runs Qwen2-VL-2B but leaves nothing spare for a longer conversation; comfortable has headroom. If you expect to grow the context, buy for comfortable.
06

Open the card you have settled on

Each card page repeats this sweep for every model we hold. It answers what else the hardware is good for, beyond Qwen2-VL-2B.

Answers

Qwen2-VL-2B — common questions

What GPU do I need to run Qwen2-VL-2B?

The smallest card in our catalogue that holds Qwen2-VL-2B is the Tesla C1080, with 4 GB of memory. It runs the model at Q8_0 using about 2.8 GB, and produces roughly 18.4 tokens per second. 818 cards in total can run it.

How fast is Qwen2-VL-2B on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 1,694 tokens per second; the slowest that still runs it manages considerably less. Reading speed is around ten tokens per second, and 789 of the cards that can run Qwen2-VL-2B clear that.

How much VRAM does Qwen2-VL-2B need?

About 2.8 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 Qwen2-VL-2B on a 8 GB GPU?

Yes. A CMP 170HX 8 GB with 8 GB runs it at Q8_0, using about 2.8 GB and generating roughly 316 tokens per second — a comfortable fit.

Can I run Qwen2-VL-2B on a 12 GB GPU?

Yes. A GeForce RTX 3080 Ti with 12 GB runs it at Q8_0, using about 2.8 GB and generating roughly 193 tokens per second — a comfortable fit.

Can I run Qwen2-VL-2B on a 16 GB GPU?

Yes. A Tesla V100 SXM2 16 GB with 16 GB runs it at Q8_0, using about 2.8 GB and generating roughly 239 tokens per second — a comfortable fit.

Can I run Qwen2-VL-2B on a 24 GB GPU?

Yes. A GeForce RTX 5090 D V2 with 24 GB runs it at Q8_0, using about 2.8 GB and generating roughly 284 tokens per second — a comfortable fit.

Is Qwen2-VL-2B open source?

Its weights are published, so Qwen2-VL-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 Qwen2-VL-2B have?

Qwen2-VL-2B has 2B parameters. 1.5 billion (language model) and 675M (vision encoder). 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-VL-2B?

Qwen2-VL-2B was published by Alibaba, based in China, categorised as industry.

When was Qwen2-VL-2B released?

Qwen2-VL-2B 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-VL-2B used for?

Qwen2-VL-2B works in Language, Vision, Multimodal, and is recorded as handling visual question answering, Video description, Language modeling/generation, Translation, Question answering, Character recognition (OCR), Quantitative reasoning. 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 Qwen2-VL-2B?

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-VL-2B?

Around 1.7 × 10²² 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-VL-2B 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 Qwen2-VL-2B is rarely worth using. Every figure here assumes the whole model is on the card.

Would two GPUs run Qwen2-VL-2B faster?

Capacity adds across cards; throughput does not. Since 818 of the cards we track already hold Qwen2-VL-2B on their own, a second card is rarely the answer here.

Why does the quantisation differ between cards for Qwen2-VL-2B?

Because capacity varies, so does how hard Qwen2-VL-2B has to be squeezed — 1 distinct levels appear in the table above. Set a minimum quality to compare at one.

How accurate are these Qwen2-VL-2B speed estimates?

They are calculated from specifications rather than measured, and each carries a range — 1,016–2,711 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.

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