YaRN (Llama 2 13B) TPS calculator

Open weights Nous Research,EleutherAI,University of Geneva 13B 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

241 cards that can run it

818 cards we hold specifications for

Smallest card that fits

Xeon Phi 7120P

16 GB · Q4_K_M · 17.2 tok/s

Fastest card

B200

261 tok/s · 180 GB

Which GPUs can run YaRN (Llama 2 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

241 cards match

Calculating

					Needs	Quantisation	Fit
261 tok/s 222–313	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	20.1 GB	Q8_0	Comfortable
261 tok/s 222–313	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	20.1 GB	Q8_0	Comfortable
208 tok/s 125–333 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	20.1 GB	Q8_0	Comfortable
208 tok/s 125–333 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	20.1 GB	Q8_0	Comfortable
166 tok/s 100–266 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	20.1 GB	Q8_0	Comfortable
159 tok/s 135–191	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	20.1 GB	Q8_0	Comfortable
159 tok/s 135–191	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	20.1 GB	Q8_0	Comfortable
152 tok/s 91–244 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	20.1 GB	Q8_0	Comfortable
135 tok/s 81–217 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	20.1 GB	Q8_0	Comfortable
135 tok/s 81–217 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	20.1 GB	Q8_0	Comfortable
135 tok/s 81–217 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	20.1 GB	Q8_0	Comfortable
128 tok/s 109–154	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	20.1 GB	Q8_0	Comfortable
109 tok/s 93–131	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	20.1 GB	Q8_0	Comfortable
109 tok/s 93–131	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	20.1 GB	Q8_0	Comfortable
109 tok/s 93–131	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	20.1 GB	Q8_0	Comfortable
109 tok/s 93–131	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	20.1 GB	Q8_0	Comfortable
109 tok/s 93–131	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	20.1 GB	Q8_0	Comfortable
85.0 tok/s 72–102	Tesla V100 SXM2 16 GB NVIDIA	16 GB	1,130 GB/s	Nov 2019	14.0 GB	Q4_K_M	Tight
83.4 tok/s 50–133 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	20.1 GB	Q8_0	Comfortable
83.4 tok/s 50–133 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	20.1 GB	Q8_0	Comfortable
72.2 tok/s 61–87	GeForce RTX 5080 NVIDIA	16 GB	960 GB/s	Jan 2025	14.0 GB	Q4_K_M	Tight
69.5 tok/s 42–111 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	20.1 GB	Q8_0	Comfortable
68.0 tok/s 41–109 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	20.1 GB	Q8_0	Comfortable
67.5 tok/s 57–81	Tesla V100 DGXS 16 GB NVIDIA	16 GB	897 GB/s	Mar 2018	14.0 GB	Q4_K_M	Tight
67.5 tok/s 57–81	Tesla V100 PCIe 16 GB NVIDIA	16 GB	897 GB/s	Jun 2017	14.0 GB	Q4_K_M	Tight

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: Nous Research,EleutherAI,University of Geneva
Organisation type: Industry,Research collective,Academia
Country: United States of America, Switzerland
Published: 1 November 2023
Authors: Bowen Peng, Jeffrey Quesnelle, Honglu Fan, Enrico Shippole

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
Base model: Llama 2-13B

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: 2,457,600,000 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: 1.9 × 10²⁰ FLOP
How it was established: Operation counting
Fine-tuning compute: 1.9 × 10²⁰ FLOP

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
Hugging Face: NousResearch

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: YaRN: Efficient Context Window Extension of Large Language Models
Last updated: 11 February 2026

The extremes

The ten fastest GPUs that run YaRN (Llama 2 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 YaRN (Llama 2 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

What you need to run it

Minimum card

Xeon Phi 7120P

Memory needed

14.0 GB

Fastest

261 tok/s

YaRN (Llama 2 13B) is small enough at 13B parameters that hardware is rarely the obstacle — 241 of the cards we track can run it, including cards several years old.

The least hardware that works is a Xeon Phi 7120P. Its 16 GB is enough at Q4_K_M compression, giving roughly 17.2 tokens per second.

At the other end, a B200 generates roughly 261 tokens per second on it, on the strength of 8,000 GB/s of memory bandwidth.

What this model is

YaRN (Llama 2 13B) was published by Nous Research,EleutherAI,University of Geneva, in United States of America, in November 2023. The organisation is categorised as industry,Research collective,Academia.

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

Its starting point was Llama 2-13B — most models at this scale are adapted from an existing base rather than built from nothing.

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 NousResearch organisation on Hugging Face.

What decides the speed

Across every card that can run it, the middle of the range is about 30.0 tokens per second, and 220 of them clear the ten tokens per second that roughly matches reading speed.

Every weight participates in every token here, so bandwidth is the whole story: the ranking below is effectively a ranking of memory throughput.

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

What went into building it

Producing it required around 1.9 × 10²⁰ FLOP of arithmetic, which is a statement about the training budget rather than about inference.

Around 2,457,600,000 tokens went into training it.

Step by step

How to choose a GPU for YaRN (Llama 2 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

Look at what YaRN (Llama 2 13B) actually needs — around 14.0 GB at Q4_K_M. No amount of processing power compensates for a card that cannot hold it.
02

Decide how long your conversations run

Longer conversations cost memory on top of what the weights need. Move the slider to your real working length before trusting any row for YaRN (Llama 2 13B).
03

Set a quality floor

Each card runs the least-compressed copy it can hold — Q4_K_M on the smallest card that fits. Setting a floor drops the cards that only manage YaRN (Llama 2 13B) by squeezing it further than you would want.
04

Compare tokens per second, not specifications

Sort by speed to see how cards rank for YaRN (Llama 2 13B). It will not match a gaming ordering — generation is bound by memory bandwidth, which is why the B200 tops it at 261 tok/s.
05

Look at the headroom, not just the fit

Tight means YaRN (Llama 2 13B) loads and works, with no room to raise the context later. Comfortable means you can. The difference matters more than a few tokens per second.
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 YaRN (Llama 2 13B) is settled.

Answers

YaRN (Llama 2 13B) — common questions

Can I run YaRN (Llama 2 13B) on a 24 GB GPU?

Yes. A GeForce RTX 5090 D V2 with 24 GB runs it at Q8_0, using about 20.1 GB and generating roughly 43.7 tokens per second — a tight fit.

Is YaRN (Llama 2 13B) open source?

Its weights are published, so YaRN (Llama 2 13B) 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 YaRN (Llama 2 13B) have?

YaRN (Llama 2 13B) has 13B parameters. same as base model. 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 YaRN (Llama 2 13B)?

YaRN (Llama 2 13B) was published by Nous Research,EleutherAI,University of Geneva, based in United States of America, categorised as industry,Research collective,Academia.

When was YaRN (Llama 2 13B) released?

YaRN (Llama 2 13B) 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 YaRN (Llama 2 13B) used for?

YaRN (Llama 2 13B) works in Language, and is recorded as handling language modeling/generation, Question answering. These are the areas it was designed around; they describe intent rather than a hard boundary.

Where can I download YaRN (Llama 2 13B)?

Its weights are published under the NousResearch 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 YaRN (Llama 2 13B)?

Around 1.9 × 10²⁰ 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 YaRN (Llama 2 13B) 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 YaRN (Llama 2 13B) is rarely worth using — the nearest miss we calculate is short by 3.2 GB. Every figure here assumes the whole model is on the card.

Would two GPUs run YaRN (Llama 2 13B) faster?

Capacity adds across cards; throughput does not. Since 241 of the cards we track already hold YaRN (Llama 2 13B) on their own, a second card is rarely the answer here.

Why does the quantisation differ between cards for YaRN (Llama 2 13B)?

A larger card holds a more accurate copy. Across the cards that run YaRN (Llama 2 13B), 3 compression levels are used; the floor control above pins it to one.

How accurate are these YaRN (Llama 2 13B) speed estimates?

They are calculated from specifications rather than measured, and each carries a range — 222–313 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 YaRN (Llama 2 13B)?

The smallest card in our catalogue that holds YaRN (Llama 2 13B) is the Xeon Phi 7120P, with 16 GB of memory. It runs the model at Q4_K_M using about 14.0 GB, and produces roughly 17.2 tokens per second. 241 cards in total can run it.

How fast is YaRN (Llama 2 13B) on a GPU?

It depends on the card. The quickest we calculate is a 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 220 of the cards that can run YaRN (Llama 2 13B) clear that.

How much VRAM does YaRN (Llama 2 13B) need?

About 14.0 GB at Q4_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 YaRN (Llama 2 13B) on a 16 GB GPU?

Yes. A Tesla V100 SXM2 16 GB with 16 GB runs it at Q4_K_M, using about 14.0 GB and generating roughly 85.0 tokens per second — a tight fit.

Source

Original publication

Record last updated 11 February 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