Stable Beluga 1 TPS calculator

Open weights Stability AI 65.2B parameters July 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

61 cards that can run it

818 cards we hold specifications for

Smallest card that fits

A100 PCIe 40 GB

40 GB · Q3_K_M · 27.3 tok/s

Fastest card

B200

52.0 tok/s · 180 GB

Which GPUs can run Stable Beluga 1?

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

61 cards match

Calculating

					Needs	Quantisation	Fit
52.0 tok/s 31–83 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	70.5 GB	Q8_0	Comfortable
52.0 tok/s 31–83 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	70.5 GB	Q8_0	Comfortable
41.5 tok/s 25–66 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	70.5 GB	Q8_0	Comfortable
41.5 tok/s 25–66 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	70.5 GB	Q8_0	Comfortable
33.2 tok/s 20–53 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	70.5 GB	Q8_0	Comfortable
31.8 tok/s 19–51 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	70.5 GB	Q8_0	Comfortable
31.8 tok/s 19–51 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	70.5 GB	Q8_0	Comfortable
30.4 tok/s 18–49 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	70.5 GB	Q8_0	Comfortable
28.0 tok/s 17–45 · low confidence	GRID A100B NVIDIA	48 GB	1,870 GB/s	May 2020	40.1 GB	Q4_K_M	Tight
27.3 tok/s 16–44 · low confidence	A100 PCIe 40 GB NVIDIA	40 GB	1,560 GB/s	Jun 2020	32.6 GB	Q3_K_M	Tight
27.3 tok/s 16–44 · low confidence	A100 SXM4 40 GB NVIDIA	40 GB	1,560 GB/s	May 2020	32.6 GB	Q3_K_M	Tight
27.3 tok/s 16–44 · low confidence	A800 PCIe 40 GB NVIDIA	40 GB	1,560 GB/s	Nov 2022	32.6 GB	Q3_K_M	Tight
27.0 tok/s 16–43 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	70.5 GB	Q8_0	Comfortable
27.0 tok/s 16–43 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	70.5 GB	Q8_0	Comfortable
27.0 tok/s 16–43 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	70.5 GB	Q8_0	Comfortable
25.6 tok/s 15–41 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	70.5 GB	Q8_0	Tight
21.8 tok/s 13–35 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	70.5 GB	Q8_0	Comfortable
21.8 tok/s 13–35 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	70.5 GB	Q8_0	Tight
21.8 tok/s 13–35 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	70.5 GB	Q8_0	Tight
21.8 tok/s 13–35 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	70.5 GB	Q8_0	Comfortable
21.8 tok/s 13–35 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	70.5 GB	Q8_0	Tight
20.1 tok/s 12–32 · low confidence	RTX PRO 5000 Blackwell NVIDIA	48 GB	1,340 GB/s	Mar 2025	40.1 GB	Q4_K_M	Tight
19.1 tok/s 11–31 · low confidence	H100 SXM5 64 GB NVIDIA	64 GB	2,020 GB/s	Mar 2023	55.3 GB	Q6_K	Tight
16.6 tok/s 10–27 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	70.5 GB	Q8_0	Comfortable
16.6 tok/s 10–27 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	70.5 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: Stability AI
Organisation type: Industry
Country: United Kingdom of Great Britain and Northern Ireland
Published: 21 July 2023

What it does

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

Domain: Language
Task: Language generation
Base model: LLaMA-65B

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: 65.2B
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 (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.

Record confidence: Confident

Sources

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

Reference: Meet Stable Beluga 1 and Stable Beluga 2, Our Large and Mighty Instruction Fine-Tuned Language Models
Last updated: 28 November 2025

The extremes

The ten fastest GPUs that run Stable Beluga 1

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 Stable Beluga 1

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 PCIe 40 GB

Memory needed

32.6 GB

Fastest

52.0 tok/s

Stable Beluga 1 reaches a parameter count of 65.2B. 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: 61.

At the low end it is handled by A100 PCIe 40 GB, with a memory capacity of 40 GB, running it at a compression of Q3_K_M and producing around 27.3 tokens per second.

Top of the range is B200, generating roughly 52.0 tokens per second on the strength of a memory bandwidth of 8,000 GB/s.

About this model

Stable Beluga 1 was published by Stability AI, in the country recorded as United Kingdom of Great Britain and Northern Ireland, during July 2023. The category the publisher falls under is industry.

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

Rather than being trained from scratch, it is derived from LLaMA-65B. That is why it shares the base model's general shape and size.

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 13.3 tokens per second. Producing text faster than most people read it: 56 of them.

Being dense, it reads all of itself per token, which is why the ordering by speed below follows the ordering by memory bandwidth so closely.

Memory here is estimated from size rather than computed from the architecture, which is not recorded for this model — the numbers are indicative rather than exact.

Step by step

How to choose a GPU for Stable Beluga 1

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

The table lists every card able to hold Stable Beluga 1, needing around 32.6 GB at a compression of Q3_K_M. That figure, not the headline performance of a card, is what decides whether it runs.
02

Set the context length you will work at

The conversation occupies memory too, and grows as it goes. Set the slider to the length you expect, because at long context a card that handles short questions easily can be dropped by Stable Beluga 1.
03

Decide how much compression you will accept

Each card runs the least-compressed copy it can hold, 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

Sort by speed to see how cards rank for Stable Beluga 1. It will not match a gaming ordering, because generation is bound by memory bandwidth. The card topping the list is B200, at 52.0 tok/s.
05

Check the fit verdict before buying

A tight fit runs, but leaves nothing spare for a longer conversation, in the case of Stable Beluga 1. 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

Check the card from the other side

Each card page repeats this sweep for every model we hold, answering what else the hardware is good for beyond Stable Beluga 1.

Answers

Stable Beluga 1 — common questions

Stable Beluga 1— how fast is it on a GPU?

It depends on the card. The quickest we calculate is B200, at about 52.0 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: 56.

Stable Beluga 1— how much VRAM does it need?

It needs about 32.6 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.

Stable Beluga 1— 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.

Stable Beluga 1— how many parameters does it have?

It has a parameter count of 65.2B. 65.2B. 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.

Stable Beluga 1— who created it?

It was published by Stability AI, based in United Kingdom of Great Britain and Northern Ireland, an organisation categorised as industry.

Stable Beluga 1— when was it released?

It was published in July 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.

Stable Beluga 1— what is it used for?

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

Stable Beluga 1— 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.

Stable Beluga 1— 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 11.3 GB. Every figure here assumes the whole model is resident on the card.

Stable Beluga 1— would two GPUs run it faster?

Two cards buy memory rather than speed, which matters only if one card cannot hold it. The number that can: 61. So a second card is rarely the answer here.

Stable Beluga 1— 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: 4. Bigger cards get the more accurate version, and the quality floor above pins the comparison to one level.

Stable Beluga 1— how accurate are these speed estimates?

They are calculated from specifications rather than measured, and each carries a range. One example: 31–83 tok/s on B200. The same model and card vary by thirty to fifty per cent depending on the inference software and its version.

Stable Beluga 1— what GPU do I need to run it?

The smallest card in our catalogue that holds it is A100 PCIe 40 GB, with a memory capacity of 40 GB. It runs the model at a compression of Q3_K_M using about 32.6 GB, and produces roughly 27.3 tokens per second. The number of cards able to run it in total: 61.

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