LLaMA-13B TPS calculator

Open weights Meta AI 13B parameters February 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

306 cards that can run it

818 cards we hold specifications for

Smallest card that fits

P102-101

10 GB · IQ4_XS · 21.8 tok/s

Fastest card

B200

261 tok/s · 180 GB

Which GPUs can run LLaMA-13B?

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

306 cards match

Calculating

					Needs	Quantisation	Fit
261 tok/s 222–313	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	15.4 GB	Q8_0	Comfortable
261 tok/s 222–313	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	15.4 GB	Q8_0	Comfortable
208 tok/s 125–333 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	15.4 GB	Q8_0	Comfortable
208 tok/s 125–333 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	15.4 GB	Q8_0	Comfortable
166 tok/s 100–266 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	15.4 GB	Q8_0	Comfortable
159 tok/s 135–191	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	15.4 GB	Q8_0	Comfortable
159 tok/s 135–191	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	15.4 GB	Q8_0	Comfortable
152 tok/s 91–244 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	15.4 GB	Q8_0	Comfortable
135 tok/s 81–217 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	15.4 GB	Q8_0	Comfortable
135 tok/s 81–217 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	15.4 GB	Q8_0	Comfortable
135 tok/s 81–217 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	15.4 GB	Q8_0	Comfortable
128 tok/s 109–154	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	15.4 GB	Q8_0	Comfortable
125 tok/s 106–150	CMP 170HX 10 GB NVIDIA	10 GB	1,560 GB/s	Sep 2021	8.6 GB	IQ4_XS	Tight
109 tok/s 93–131	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	15.4 GB	Q8_0	Comfortable
109 tok/s 93–131	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	15.4 GB	Q8_0	Comfortable
109 tok/s 93–131	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	15.4 GB	Q8_0	Comfortable
109 tok/s 93–131	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	15.4 GB	Q8_0	Comfortable
109 tok/s 93–131	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	15.4 GB	Q8_0	Comfortable
83.4 tok/s 50–133 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	15.4 GB	Q8_0	Comfortable
83.4 tok/s 50–133 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	15.4 GB	Q8_0	Comfortable
69.5 tok/s 42–111 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	15.4 GB	Q8_0	Comfortable
68.6 tok/s 58–82	GeForce RTX 3080 12 GB NVIDIA	12 GB	912 GB/s	Jan 2022	9.3 GB	Q4_K_M	Tight
68.6 tok/s 58–82	GeForce RTX 3080 Ti NVIDIA	12 GB	912 GB/s	May 2021	9.3 GB	Q4_K_M	Tight
68.0 tok/s 41–109 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	15.4 GB	Q8_0	Comfortable
66.5 tok/s 56–80	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	15.4 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: Meta AI
Organisation type: Industry
Country: United States of America
Published: 27 February 2023
Authors: Hugo Touvron, Thibaut Lavril, Gautier Izacard, Xavier Martinet, Marie-Anne Lachaux, Timothée Lacroix, Baptiste Rozière, Naman Goyal, Eric Hambro, Faisal Azhar, Aurelien Rodriguez, Armand Joulin, Edouard Grave, Guillaume Lample

What it does

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

Domain: Language
Task: Language modeling, Code generation

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: 13B
Training data: 1,000,000,000,000 tokens
Epochs: 1
Batch size: 4,000,000

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: 7.8 × 10²² FLOP
How it was established: Operation counting

The training run

What it physically took to train: which chips, how many, for how long, and what that drew from the wall.

Training hardware: NVIDIA A100
Hardware utilisation: MFU 51.4%
Compute cost: $61,301

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

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: Highly cited
Record confidence: Confident
Citations: 19,926
Benchmark data: LLaMA-13B

Sources

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

Reference: LLaMA: Open and Efficient Foundation Language Models
Last updated: 25 May 2026

The extremes

The ten fastest GPUs that run LLaMA-13B

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 LLaMA-13B

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

P102-101

Memory needed

8.6 GB

Fastest

261 tok/s

LLaMA-13B reaches a parameter count of 13B. That is small enough that hardware is rarely the obstacle, including on cards several years old. The number of cards we track that can run it: 306.

At the low end it is handled by P102-101, with a memory capacity of 10 GB, running it at a compression of IQ4_XS and producing around 21.8 tokens per second.

The fastest we calculate for it is B200, generating roughly 261 tokens per second on the strength of a memory bandwidth of 8,000 GB/s.

About this model

LLaMA-13B was published by Meta AI, in the country recorded as United States of America, during February 2023. It comes out of an organisation categorised as industry.

It works in the domain of Language, and is recorded as performing the task of language modeling, Code generation.

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.

How fast it runs, and why

Across every card that can run it, the middle of the range sits at 24.3 tokens per second. Clearing the ten tokens per second that roughly matches reading speed: 269 of them.

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.

Its attention layout is on file, so the memory figures are computed exactly rather than approximated.

Training and provenance

Training it took a computation budget of roughly 7.8 × 10²² FLOP, on hardware recorded as NVIDIA A100. That figure measures what producing the model cost, and has no bearing on how fast it answers.

Training consumed a corpus of around 1,000,000,000,000 tokens of text.

The reason it appears in this catalogue at all: highly cited.

Step by step

How to choose a GPU for LLaMA-13B

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 LLaMA-13B, needing around 8.6 GB at a compression of IQ4_XS. No amount of processing power compensates for a card that cannot hold it.
02

Set the context length you will work at

Set the context to what you will actually use. The cache grows with the conversation, and it is the usual reason a card that seemed fine stops fitting LLaMA-13B.
03

Set a quality floor

Each card runs the least-compressed copy it can hold, reaching a compression of IQ4_XS on the smallest card that fits. Setting a minimum quality drops the cards that only manage it by squeezing further than you would want, and holds the comparison at one level.
04

Rank by throughput rather than spec sheet

Sort by speed to see how cards rank for LLaMA-13B. It will not match a gaming ordering, because generation is bound by memory bandwidth. The card topping the list is B200, at 261 tok/s.
05

Read the fit column last

Tight means it loads and works with no room to raise the context later, in the case of LLaMA-13B. Comfortable means you can grow the context later. That difference matters more than a few tokens per second, so buy for comfortable if you expect to.
06

See what else that card runs

Following a card through to its own page shows every other model it can hold, which is the question that follows once you have settled on LLaMA-13B.

Answers

LLaMA-13B — common questions

LLaMA-13B— why does the quantisation differ between cards?

Because capacity varies, so does how hard it has to be squeezed. The number of distinct levels in the table above: 4. Bigger cards get the more accurate version, and the quality floor above pins the comparison to one level.

LLaMA-13B— how accurate are these speed estimates?

They are calculated from specifications rather than measured, and each carries a range. One example: 222–313 tok/s on B200. The same model and card vary by thirty to fifty per cent depending on the inference software and its version.

LLaMA-13B— what GPU do I need to run it?

The smallest card in our catalogue that holds it is P102-101, with a memory capacity of 10 GB. It runs the model at a compression of IQ4_XS using about 8.6 GB, and produces roughly 21.8 tokens per second. The number of cards able to run it in total: 306.

LLaMA-13B— how fast is it on a GPU?

It depends on the card. The quickest we calculate is B200, at about 261 tokens per second; the slowest that still runs it manages considerably less. Reading speed is around ten tokens per second, and the number of cards clearing that: 269.

LLaMA-13B— how much VRAM does it need?

It needs about 8.6 GB at a compression of IQ4_XS, 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.

LLaMA-13B— can I run it on a GPU holding 12 GB?

Yes. The card GeForce RTX 3080 Ti, holding 12 GB, runs it at a compression of Q4_K_M, using about 9.3 GB and generating roughly 68.6 tokens per second. The fit is tight.

LLaMA-13B— can I run it on a GPU holding 16 GB?

Yes. The card Tesla V100 SXM2 16 GB, holding 16 GB, runs it at a compression of Q6_K, using about 12.3 GB and generating roughly 53.5 tokens per second. The fit is tight.

LLaMA-13B— can I run it on a GPU holding 24 GB?

Yes. The card GeForce RTX 5090 D V2, holding 24 GB, runs it at a compression of Q8_0, using about 15.4 GB and generating roughly 43.7 tokens per second. The fit is comfortable.

LLaMA-13B— is it open source?

Its weights are published, so it 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.

LLaMA-13B— how many parameters does it have?

It has a parameter count of 13B. 13.0B. 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.

LLaMA-13B— who created it?

It was published by Meta AI, based in United States of America, an organisation categorised as industry.

LLaMA-13B— when was it released?

It was published in February 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.

LLaMA-13B— what is it used for?

It works in the domain of Language, and is recorded as handling the task of language modeling, Code generation. Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.

LLaMA-13B— where can I download it?

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

LLaMA-13B— how much compute was used to train it?

Training consumed around 7.8 × 10²² FLOP, on hardware recorded as NVIDIA A100. 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.

LLaMA-13B— can I run it 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 model is rarely worth using. The nearest miss we calculate falls short by 2.1 GB. Every figure here assumes the whole model is resident on the card.

LLaMA-13B— would two GPUs run it faster?

A second card roughly doubles the memory available but not the generation rate. The number already able to run it alone: 306. So a second card is rarely the answer here.

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