Vicuna-33B-v1.3 TPS calculator

Open weights Large Model Systems Organization,University of California (UC) Berkeley 33B parameters June 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

132 cards that can run it

818 cards we hold specifications for

Smallest card that fits

RTX A4500

20 GB · Q3_K_M · 22.2 tok/s

Fastest card

B200

103 tok/s · 180 GB

Which GPUs can run Vicuna-33B-v1.3?

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

132 cards match

Calculating

					Needs	Quantisation	Fit
103 tok/s 62–164 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	36.0 GB	Q8_0	Comfortable
103 tok/s 62–164 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	36.0 GB	Q8_0	Comfortable
82.0 tok/s 49–131 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	36.0 GB	Q8_0	Comfortable
82.0 tok/s 49–131 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	36.0 GB	Q8_0	Comfortable
65.6 tok/s 39–105 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	36.0 GB	Q8_0	Comfortable
62.8 tok/s 38–100 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	36.0 GB	Q8_0	Comfortable
62.8 tok/s 38–100 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	36.0 GB	Q8_0	Comfortable
60.1 tok/s 36–96 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	36.0 GB	Q8_0	Comfortable
53.3 tok/s 32–85 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	36.0 GB	Q8_0	Comfortable
53.3 tok/s 32–85 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	36.0 GB	Q8_0	Comfortable
53.3 tok/s 32–85 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	36.0 GB	Q8_0	Comfortable
50.6 tok/s 30–81 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	36.0 GB	Q8_0	Comfortable
43.1 tok/s 26–69 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	36.0 GB	Q8_0	Comfortable
43.1 tok/s 26–69 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	36.0 GB	Q8_0	Comfortable
43.1 tok/s 26–69 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	36.0 GB	Q8_0	Comfortable
43.1 tok/s 26–69 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	36.0 GB	Q8_0	Comfortable
43.1 tok/s 26–69 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	36.0 GB	Q8_0	Comfortable
39.7 tok/s 24–64 · low confidence	GeForce RTX 5090 D V2 NVIDIA	24 GB	1,340 GB/s	Aug 2025	20.7 GB	Q4_K_M	Tight
36.2 tok/s 22–58 · low confidence	A30X NVIDIA	24 GB	1,220 GB/s	Apr 2021	20.7 GB	Q4_K_M	Tight
34.9 tok/s 21–56 · low confidence	DRIVE A100 PROD NVIDIA	32 GB	1,870 GB/s	May 2020	28.3 GB	Q6_K	Tight
34.9 tok/s 21–56 · low confidence	GRID A100A NVIDIA	32 GB	1,870 GB/s	May 2020	28.3 GB	Q6_K	Tight
33.4 tok/s 20–53 · low confidence	GeForce RTX 5090 NVIDIA	32 GB	1,790 GB/s	Jan 2025	28.3 GB	Q6_K	Tight
33.4 tok/s 20–53 · low confidence	GeForce RTX 5090 D NVIDIA	32 GB	1,790 GB/s	Jan 2025	28.3 GB	Q6_K	Tight
32.8 tok/s 20–53 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	36.0 GB	Q8_0	Comfortable
32.8 tok/s 20–53 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	36.0 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: Large Model Systems Organization,University of California (UC) Berkeley
Organisation type: Academia,Academia
Country: United States of America
Published: 22 June 2023

What it does

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

Domain: Language
Task: Language modeling/generation, Chat
Base model: LLaMA-33B

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: 33B
Training data: 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.

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 (restricted use)
Training code: Open source

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: Vicuna Model Card
Last updated: 28 November 2025

The extremes

The ten fastest GPUs that run Vicuna-33B-v1.3

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 Vicuna-33B-v1.3

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

RTX A4500

Memory needed

16.8 GB

Fastest

103 tok/s

With 33B parameters, Vicuna-33B-v1.3 lands in the range a serious desktop card can handle once the weights are compressed. 132 of the cards we track can run it.

The least hardware that works is a RTX A4500. Its 20 GB is enough at Q3_K_M compression, giving roughly 22.2 tokens per second.

Top of the range is the B200, at roughly 103 tokens per second thanks to 8,000 GB/s of bandwidth.

Background

Vicuna-33B-v1.3 was published by Large Model Systems Organization,University of California (UC) Berkeley, in United States of America, in June 2023. It comes out of academia,Academia.

It works in Language, and is recorded as doing language modeling/generation, Chat.

It builds on LLaMA-33B, which is why it shares that model's general shape and size.

The weights are published, so it can be downloaded and run on your own hardware indefinitely, offline, with no account attached.

Reading the throughput figures

Half the cards that hold it manage more than 20.3 tokens per second, and 101 exceed reading speed outright.

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.

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.

Step by step

How to choose a GPU for Vicuna-33B-v1.3

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 Vicuna-33B-v1.3 — around 16.8 GB at Q3_K_M. Capacity is the gate — a card either holds it or it does not.
02

Match the context to your actual use

Set the context to what you will actually use. The cache grows with the conversation, and it is the usual reason Vicuna-33B-v1.3 stops fitting a card that seemed fine.
03

Choose how far you will compress it

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 Vicuna-33B-v1.3 by squeezing it further than you would want.
04

Sort by speed

Sort by speed to see how cards rank for Vicuna-33B-v1.3. It will not match a gaming ordering — generation is bound by memory bandwidth, which is why the B200 tops it at 103 tok/s.
05

Check the fit verdict before buying

A tight fit runs Vicuna-33B-v1.3 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

Following a card through to its own page shows every other model it can hold, which is the question that follows once Vicuna-33B-v1.3 is settled.

Answers

Vicuna-33B-v1.3 — common questions

Can I run Vicuna-33B-v1.3 on a 24 GB GPU?

Yes. A GeForce RTX 5090 D V2 with 24 GB runs it at Q4_K_M, using about 20.7 GB and generating roughly 39.7 tokens per second — a tight fit.

Is Vicuna-33B-v1.3 open source?

Its weights are published, so Vicuna-33B-v1.3 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 Vicuna-33B-v1.3 have?

Vicuna-33B-v1.3 has 33B 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 Vicuna-33B-v1.3?

Vicuna-33B-v1.3 was published by Large Model Systems Organization,University of California (UC) Berkeley, based in United States of America, categorised as academia,Academia.

When was Vicuna-33B-v1.3 released?

Vicuna-33B-v1.3 was published in June 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 Vicuna-33B-v1.3 used for?

Vicuna-33B-v1.3 works in Language, and is recorded as handling language modeling/generation, Chat. These are the areas it was designed around; they describe intent rather than a hard boundary.

Where can I download Vicuna-33B-v1.3?

The weights for Vicuna-33B-v1.3 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 Vicuna-33B-v1.3 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 Vicuna-33B-v1.3 is rarely worth using — the nearest miss we calculate is short by 6.3 GB. Every figure here assumes the whole model is on the card.

Would two GPUs run Vicuna-33B-v1.3 faster?

A second card roughly doubles the memory available but not the generation rate. With 132 cards already able to run Vicuna-33B-v1.3 alone, the case for pairing is weak.

Why does the quantisation differ between cards for Vicuna-33B-v1.3?

Because capacity varies, so does how hard Vicuna-33B-v1.3 has to be squeezed — 5 distinct levels appear in the table above. Set a minimum quality to compare at one.

How accurate are these Vicuna-33B-v1.3 speed estimates?

They are calculated from specifications rather than measured, and each carries a range — 62–164 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 Vicuna-33B-v1.3?

The smallest card in our catalogue that holds Vicuna-33B-v1.3 is the RTX A4500, with 20 GB of memory. It runs the model at Q3_K_M using about 16.8 GB, and produces roughly 22.2 tokens per second. 132 cards in total can run it.

How fast is Vicuna-33B-v1.3 on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 103 tokens per second; the slowest that still runs it manages considerably less. Reading speed is around ten tokens per second, and 101 of the cards that can run Vicuna-33B-v1.3 clear that.

How much VRAM does Vicuna-33B-v1.3 need?

About 16.8 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.

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