YaLM TPS calculator

Open weights Yandex 100B parameters June 2022

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

43 cards that can run it

818 cards we hold specifications for

Smallest card that fits

Radeon Instinct MI200

64 GB · IQ4_XS · 13.3 tok/s

Fastest card

B200

33.9 tok/s · 180 GB

Which GPUs can run YaLM?

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

43 cards match

Calculating

					Needs	Quantisation	Fit
33.9 tok/s 20–54 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	107.8 GB	Q8_0	Comfortable
33.9 tok/s 20–54 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	107.8 GB	Q8_0	Comfortable
32.9 tok/s 20–53 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	61.2 GB	Q4_K_M	Tight
32.9 tok/s 20–53 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	61.2 GB	Q4_K_M	Tight
27.1 tok/s 16–43 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	107.8 GB	Q8_0	Comfortable
27.1 tok/s 16–43 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	107.8 GB	Q8_0	Comfortable
24.3 tok/s 15–39 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	84.5 GB	Q6_K	Tight
21.6 tok/s 13–35 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	107.8 GB	Q8_0	Tight
21.0 tok/s 13–34 · low confidence	H100 SXM5 64 GB NVIDIA	64 GB	2,020 GB/s	Mar 2023	55.4 GB	IQ4_XS	Tight
20.7 tok/s 12–33 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	107.8 GB	Q8_0	Tight
20.7 tok/s 12–33 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	107.8 GB	Q8_0	Tight
20.7 tok/s 12–33 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	84.5 GB	Q6_K	Tight
20.7 tok/s 12–33 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	84.5 GB	Q6_K	Tight
20.7 tok/s 12–33 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	84.5 GB	Q6_K	Tight
20.0 tok/s 12–32 · low confidence	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	61.2 GB	Q4_K_M	Tight
20.0 tok/s 12–32 · low confidence	A100X NVIDIA	80 GB	2,040 GB/s	Jun 2021	61.2 GB	Q4_K_M	Tight
20.0 tok/s 12–32 · low confidence	A800 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Aug 2022	61.2 GB	Q4_K_M	Tight
20.0 tok/s 12–32 · low confidence	H100 CNX NVIDIA	80 GB	2,040 GB/s	Mar 2023	61.2 GB	Q4_K_M	Tight
20.0 tok/s 12–32 · low confidence	H100 PCIe 80 GB NVIDIA	80 GB	2,040 GB/s	Oct 2022	61.2 GB	Q4_K_M	Tight
20.0 tok/s 12–32 · low confidence	H800 PCIe 80 GB NVIDIA	80 GB	2,040 GB/s	Mar 2023	61.2 GB	Q4_K_M	Tight
19.8 tok/s 12–32 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	107.8 GB	Q8_0	Comfortable
19.0 tok/s 11–30 · low confidence	A100 PCIe 80 GB NVIDIA	80 GB	1,940 GB/s	Jun 2021	61.2 GB	Q4_K_M	Tight
19.0 tok/s 11–30 · low confidence	A800 PCIe 80 GB NVIDIA	80 GB	1,940 GB/s	Nov 2022	61.2 GB	Q4_K_M	Tight
17.6 tok/s 11–28 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	107.8 GB	Q8_0	Tight
17.6 tok/s 11–28 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	107.8 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: Yandex
Organisation type: Industry
Country: Russia
Published: 23 June 2022
Authors: Mikhail Khrushchev, Ruslan Vasilev, Alexey Petrov, Nikolay Zinov

What it does

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

Domain: Language
Task: Language modeling, Chat
Approach: Self-supervised learning

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: 100B
Training data: 300,000,000,000 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: 2.2 × 10²³ FLOP
How it was established: Hardware

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
Chips used: 800
Chip-hours: 1,248,000
Wall-clock time: 1,560 hours (65 days)
Power draw: 642.0 kW

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

How it is classified

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

Foundation model: Yes
Likely above 10²³ FLOP: Yes
Record confidence: Likely

Sources

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

Reference: Yandex Publishes YaLM 100B. It’s the Largest GPT-Like Neural Network in Open Source
Last updated: 28 November 2025

The extremes

The ten fastest GPUs that run YaLM

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 YaLM

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

Radeon Instinct MI200

Memory needed

55.4 GB

Fastest

33.9 tok/s

YaLM reaches a parameter count of 100B. That puts it above consumer hardware, into the range where a card is bought for this purpose rather than repurposed for it. The number of cards we track that can hold it: 43.

The entry point is Radeon Instinct MI200, with a memory capacity of 64 GB, running it at a compression of IQ4_XS and producing around 13.3 tokens per second.

The fastest we calculate for it is B200, generating roughly 33.9 tokens per second on the strength of a memory bandwidth of 8,000 GB/s.

Where it came from

YaLM was published by Yandex, in the country recorded as Russia, during June 2022. The publishing organisation is categorised as industry.

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

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.

Understanding the speeds

Across every card that can run it, the middle of the range sits at 19.0 tokens per second. Clearing the ten tokens per second that roughly matches reading speed: 36 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.

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.

How it was trained

Producing it required arithmetic totalling around 2.2 × 10²³ 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 300,000,000,000 tokens of text.

Step by step

How to choose a GPU for YaLM

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 YaLM, needing around 55.4 GB at a compression of IQ4_XS. Capacity is the gate — a card either holds it or it does not.
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 YaLM.
03

Decide how much compression you will accept

Compression is what makes a model fit smaller cards, at some cost in accuracy, 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

Compare tokens per second, not specifications

Ranking by tokens per second follows memory bandwidth rather than core counts, for YaLM. It will not match a gaming ordering, because generation is bound by memory bandwidth. The card topping the list is B200, at 33.9 tok/s.
05

Check the fit verdict before buying

The fit column separates cards that just manage it from those with room to spare, in the case of YaLM. 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

Every card name links to its own page, which runs the same calculation across the whole model catalogue. A card is usually bought for more than one model, so it is worth a look before buying for YaLM.

Answers

YaLM — common questions

YaLM— 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: 43. So a second card is rarely the answer here.

YaLM— why does the quantisation differ between cards?

Each card is shown running the least-compressed copy it can hold. The number of distinct compression levels across the cards that fit it: 5. Bigger cards get the more accurate version, and the quality floor above pins the comparison to one level.

YaLM— how accurate are these speed estimates?

These are estimates with real error bars, and any of them could reasonably land anywhere in its published range depending on which runtime you use. The fastest result here: 20–54 tok/s on B200. The same model and card vary by thirty to fifty per cent depending on the inference software and its version.

YaLM— what GPU do I need to run it?

The smallest card in our catalogue that holds it is Radeon Instinct MI200, with a memory capacity of 64 GB. It runs the model at a compression of IQ4_XS using about 55.4 GB, and produces roughly 13.3 tokens per second. The number of cards able to run it in total: 43.

YaLM— how fast is it on a GPU?

It depends on the card. The quickest we calculate is B200, at about 33.9 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: 36.

YaLM— how much VRAM does it need?

It needs about 55.4 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.

YaLM— 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.

YaLM— how many parameters does it have?

It has a parameter count of 100B. 100B. 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.

YaLM— who created it?

It was published by Yandex, based in Russia, an organisation categorised as industry.

YaLM— when was it released?

It was published in June 2022. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.

YaLM— what is it used for?

It works in the domain of Language, and is recorded as handling the task of language modeling, Chat. These are the areas it was designed around; they describe intent rather than a hard boundary.

YaLM— 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.

YaLM— how much compute was used to train it?

Training consumed around 2.2 × 10²³ 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.

YaLM— can I run it if it does not fit in my GPU?

It can be split between the card and system memory, but it generates painfully slowly that way. The nearest miss we calculate falls short by 18.0 GB. Every figure here assumes the whole model is resident on the card.

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