xLAM-8x22B TPS calculator

Open weights Salesforce 141B 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

37 cards that can run it

818 cards we hold specifications for

Smallest card that fits

A100 SXM4 80 GB

80 GB · Q3_K_M · 16.5 tok/s

Fastest card

H100 NVL 94 GB

29.1 tok/s · 94 GB

Which GPUs can run xLAM-8x22B?

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

37 cards match

Calculating

					Needs	Quantisation	Fit
29.1 tok/s 17–47 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	77.8 GB	IQ4_XS	Tight
27.4 tok/s 16–44 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	102.4 GB	Q5_K_M	Tight
27.2 tok/s 16–44 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	69.6 GB	Q3_K_M	Tight
27.2 tok/s 16–44 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	69.6 GB	Q3_K_M	Tight
24.8 tok/s 15–40 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	77.8 GB	IQ4_XS	Tight
24.0 tok/s 14–38 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	151.7 GB	Q8_0	Tight
24.0 tok/s 14–38 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	151.7 GB	Q8_0	Comfortable
23.3 tok/s 14–37 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	86.0 GB	Q4_K_M	Tight
23.3 tok/s 14–37 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	86.0 GB	Q4_K_M	Tight
22.3 tok/s 13–36 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	102.4 GB	Q5_K_M	Tight
21.3 tok/s 13–34 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	118.8 GB	Q6_K	Tight
21.3 tok/s 13–34 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	118.8 GB	Q6_K	Tight
19.2 tok/s 12–31 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	151.7 GB	Q8_0	Comfortable
19.2 tok/s 12–31 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	151.7 GB	Q8_0	Comfortable
16.5 tok/s 10–26 · low confidence	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	69.6 GB	Q3_K_M	Tight
16.5 tok/s 10–26 · low confidence	A100X NVIDIA	80 GB	2,040 GB/s	Jun 2021	69.6 GB	Q3_K_M	Tight
16.5 tok/s 10–26 · low confidence	A800 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Aug 2022	69.6 GB	Q3_K_M	Tight
16.5 tok/s 10–26 · low confidence	H100 CNX NVIDIA	80 GB	2,040 GB/s	Mar 2023	69.6 GB	Q3_K_M	Tight
16.5 tok/s 10–26 · low confidence	H100 PCIe 80 GB NVIDIA	80 GB	2,040 GB/s	Oct 2022	69.6 GB	Q3_K_M	Tight
16.5 tok/s 10–26 · low confidence	H800 PCIe 80 GB NVIDIA	80 GB	2,040 GB/s	Mar 2023	69.6 GB	Q3_K_M	Tight
15.7 tok/s 9–25 · low confidence	A100 PCIe 80 GB NVIDIA	80 GB	1,940 GB/s	Jun 2021	69.6 GB	Q3_K_M	Tight
15.7 tok/s 9–25 · low confidence	A800 PCIe 80 GB NVIDIA	80 GB	1,940 GB/s	Nov 2022	69.6 GB	Q3_K_M	Tight
14.1 tok/s 8–22 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	151.7 GB	Q8_0	Comfortable
13.7 tok/s 8–22 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	102.4 GB	Q5_K_M	Tight
13.7 tok/s 8–22 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	102.4 GB	Q5_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: Salesforce
Organisation type: Industry
Country: United States of America
Published: 6 September 2024
Authors: Jianguo Zhang, Tian Lan, Ming Zhu, Zuxin Liu, Thai Hoang, Shirley Kokane, Weiran Yao, Juntao Tan, Akshara Prabhakar, Zhiwei Liu, Haolin Chen, Yihao Feng,Tulika Awalgaonkar, Rithesh Murthy, Eric Hu, Zeyuan Chen, Ran Xu, Juan Carlos Niebles, Shelby Heinecke, Huan Wang, Silvio Savarese, Caiming Xiong

What it does

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

Domain: Language
Task: Language modeling/generation
Approach: Supervised
Base model: Mixtral 8x22B

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: 141B
Training data: tokens

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)

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: Actions Speak Louder Than Words: Introducing xLAM, Salesforce’s family of Large Action Models
Last updated: 28 November 2025

The extremes

The ten fastest GPUs that run xLAM-8x22B

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 xLAM-8x22B

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

A100 SXM4 80 GB

Memory needed

69.6 GB

Fastest

29.1 tok/s

xLAM-8x22B reaches a parameter count of 141B. 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: 37.

At the low end it is handled by A100 SXM4 80 GB, with a memory capacity of 80 GB, running it at a compression of Q3_K_M and producing around 16.5 tokens per second.

The fastest we calculate for it is H100 NVL 94 GB, generating roughly 29.1 tokens per second on the strength of a memory bandwidth of 3,940 GB/s.

Where it came from

xLAM-8x22B was published by Salesforce, in the country recorded as United States of America, during September 2024. The publishing organisation is categorised as industry.

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

Its starting point was an existing base model, Mixtral 8x22B. That is why it shares the base model's general shape and size.

Because its weights were released, nothing about running it depends on a provider staying available — it is yours once downloaded.

Understanding the speeds

Across every card that can run it, the middle of the range sits at 16.5 tokens per second. Producing text faster than most people read it: 35 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.

Step by step

How to choose a GPU for xLAM-8x22B

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

01

Check what it needs before anything else

Start from what it actually needs, which is the requirement of xLAM-8x22B, needing around 69.6 GB at a compression of Q3_K_M. No amount of processing power compensates for a card that cannot hold it.
02

Decide how long your conversations run

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 xLAM-8x22B.
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 Q3_K_M 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

The speed ordering is effectively an ordering by memory bandwidth, for xLAM-8x22B. It will not match a gaming ordering, because generation is bound by memory bandwidth. The card topping the list is H100 NVL 94 GB, at 29.1 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 xLAM-8x22B. 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

Open the card you have settled on

Each card page repeats this sweep for every model we hold, answering what else the hardware is good for beyond xLAM-8x22B.

Answers

xLAM-8x22B — common questions

xLAM-8x22B— 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.

xLAM-8x22B— can I run it if it does not fit in my GPU?

Only by offloading, which is usually a false economy: the part held in system memory drags the whole thing down. The nearest miss we calculate falls short by 21.2 GB. Every figure here assumes the whole model is resident on the card.

xLAM-8x22B— would two GPUs run it faster?

Capacity adds across cards; throughput does not. The number of cards already holding it on their own: 37. So a second card is rarely the answer here.

xLAM-8x22B— why does the quantisation differ between cards?

A larger card holds a more accurate copy. The number of compression levels used across the cards that run it: 6. Bigger cards get the more accurate version, and the quality floor above pins the comparison to one level.

xLAM-8x22B— how accurate are these speed estimates?

They are calculated from specifications rather than measured, and each carries a range. One example: 17–47 tok/s on H100 NVL 94 GB. The same model and card vary by thirty to fifty per cent depending on the inference software and its version.

xLAM-8x22B— what GPU do I need to run it?

The smallest card in our catalogue that holds it is A100 SXM4 80 GB, with a memory capacity of 80 GB. It runs the model at a compression of Q3_K_M using about 69.6 GB, and produces roughly 16.5 tokens per second. The number of cards able to run it in total: 37.

xLAM-8x22B— how fast is it on a GPU?

It depends on the card. The quickest we calculate is H100 NVL 94 GB, at about 29.1 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: 35.

xLAM-8x22B— how much VRAM does it need?

It needs about 69.6 GB at a compression of Q3_K_M, 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.

xLAM-8x22B— 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.

xLAM-8x22B— how many parameters does it have?

It has a parameter count of 141B. 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.

xLAM-8x22B— who created it?

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

xLAM-8x22B— when was it released?

It 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.

xLAM-8x22B— what is it used for?

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

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