GPT-1 TPS calculator

Open weights OpenAI 117M parameters June 2018

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

818 cards that can run it

818 cards we hold specifications for

Smallest card that fits

Tesla C1080

4 GB · Q8_0 · 315 tok/s

Fastest card

B200

28,959 tok/s · 180 GB

Which GPUs can run GPT-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

818 cards match

Calculating

					Needs	Quantisation	Fit
28,959 tok/s 17,376–46,335 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	0.8 GB	Q8_0	Comfortable
28,959 tok/s 17,376–46,335 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	0.8 GB	Q8_0	Comfortable
23,125 tok/s 13,875–37,000 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	0.8 GB	Q8_0	Comfortable
23,125 tok/s 13,875–37,000 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	0.8 GB	Q8_0	Comfortable
18,494 tok/s 11,096–29,591 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	0.8 GB	Q8_0	Comfortable
17,701 tok/s 10,621–28,322 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	0.8 GB	Q8_0	Comfortable
17,701 tok/s 10,621–28,322 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	0.8 GB	Q8_0	Comfortable
16,941 tok/s 10,165–27,106 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	0.8 GB	Q8_0	Comfortable
15,035 tok/s 9,021–24,056 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	0.8 GB	Q8_0	Comfortable
15,035 tok/s 9,021–24,056 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	0.8 GB	Q8_0	Comfortable
15,035 tok/s 9,021–24,056 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	0.8 GB	Q8_0	Comfortable
14,262 tok/s 8,557–22,820 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	0.8 GB	Q8_0	Comfortable
12,163 tok/s 7,298–19,461 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	0.8 GB	Q8_0	Comfortable
12,163 tok/s 7,298–19,461 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	0.8 GB	Q8_0	Comfortable
12,163 tok/s 7,298–19,461 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	0.8 GB	Q8_0	Comfortable
12,163 tok/s 7,298–19,461 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	0.8 GB	Q8_0	Comfortable
12,163 tok/s 7,298–19,461 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	0.8 GB	Q8_0	Comfortable
9,261 tok/s 5,557–14,818 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	0.8 GB	Q8_0	Comfortable
9,261 tok/s 5,557–14,818 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	0.8 GB	Q8_0	Comfortable
7,718 tok/s 4,631–12,348 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	0.8 GB	Q8_0	Comfortable
7,553 tok/s 4,532–12,085 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	0.8 GB	Q8_0	Comfortable
7,385 tok/s 4,431–11,815 · low confidence	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	0.8 GB	Q8_0	Comfortable
7,385 tok/s 4,431–11,815 · low confidence	A100X NVIDIA	80 GB	2,040 GB/s	Jun 2021	0.8 GB	Q8_0	Comfortable
7,385 tok/s 4,431–11,815 · low confidence	A800 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Aug 2022	0.8 GB	Q8_0	Comfortable
7,385 tok/s 4,431–11,815 · low confidence	H100 CNX NVIDIA	80 GB	2,040 GB/s	Mar 2023	0.8 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: OpenAI
Organisation type: Industry
Country: United States of America
Published: 1 June 2018
Authors: A Radford, K Narasimhan, T Salimans, I Sutskever

What it does

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

Domain: Language
Task: Question answering, Text classification, Language modeling
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: 117M
Training data: 1,333,333,333 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: 1.8 × 10¹⁹ FLOP
How it was established: Operation counting,Third-party estimation

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 Quadro P600
Chips used: 8
Wall-clock time: 720 hours (30 days)
Power draw: 664 W

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: Open source

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: Likely
Citations: 13,799

Sources

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

Reference: Improving Language Understanding by Generative Pre-Training
Last updated: 1 January 2026

The extremes

The ten fastest GPUs that run GPT-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 GPT-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

Tesla C1080

Memory needed

0.8 GB

Fastest

28,959 tok/s

GPT-1 is small enough at 117M parameters that hardware is rarely the obstacle — 818 of the cards we track can run it, including cards several years old.

At the low end, a Tesla C1080 handles it — 4 GB, at Q8_0, for about 315 tokens per second.

The quickest result comes from a B200 at around 28,959 tokens per second — its 8,000 GB/s of bandwidth is what buys that.

Background

GPT-1 was published by OpenAI, in United States of America, in June 2018. industry is the category the publisher falls under.

It works in Language, and is recorded as doing question answering, Text classification, Language modeling.

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.

Reading the throughput figures

Half the cards that hold it manage more than 813.2 tokens per second, and 818 exceed reading speed outright.

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.

Training and provenance

Training it took roughly 1.8 × 10¹⁹ FLOP of computation, on NVIDIA Quadro P600 — a measure of what producing the model cost, not of how fast it answers.

It was trained on about 1,333,333,333 tokens of text.

Its inclusion criterion is highly cited.

Step by step

How to choose a GPU for GPT-1

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

Every card here has been checked against GPT-1 — around 0.8 GB at Q8_0. 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 GPT-1.
03

Set a quality floor

Compression is what makes GPT-1 fit smaller cards, at some cost in accuracy — Q8_0 on the smallest card that fits. A minimum quality removes the ones that go too far.
04

Compare tokens per second, not specifications

Ranking by tokens per second for GPT-1 follows memory bandwidth, not core counts, which is why the B200 tops it at 28,959 tok/s.
05

Look at the headroom, not just the fit

A tight fit runs GPT-1 but leaves nothing spare for a longer conversation; comfortable has headroom. If you expect to grow the context, buy for comfortable.
06

Check the card from the other side

Every card name links to its own page, which runs the same calculation across the whole model catalogue. Worth a look before buying for GPT-1 alone — a card is usually bought for more than one model.

Answers

GPT-1 — common questions

Can I run GPT-1 on a 8 GB GPU?

Yes. A CMP 170HX 8 GB with 8 GB runs it at Q8_0, using about 0.8 GB and generating roughly 5,394 tokens per second — a comfortable fit.

Can I run GPT-1 on a 12 GB GPU?

Yes. A GeForce RTX 3080 Ti with 12 GB runs it at Q8_0, using about 0.8 GB and generating roughly 3,303 tokens per second — a comfortable fit.

Can I run GPT-1 on a 16 GB GPU?

Yes. A Tesla V100 SXM2 16 GB with 16 GB runs it at Q8_0, using about 0.8 GB and generating roughly 4,091 tokens per second — a comfortable fit.

Can I run GPT-1 on a 24 GB GPU?

Yes. A GeForce RTX 5090 D V2 with 24 GB runs it at Q8_0, using about 0.8 GB and generating roughly 4,851 tokens per second — a comfortable fit.

Is GPT-1 open source?

Its weights are published, so GPT-1 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 GPT-1 have?

GPT-1 has 117M parameters. "The model had 117M parameters in total." source: https://medium.com/walmartglobaltech/the-journey-of-open-ai-gpt-models-32d95b7b7fb2. 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 GPT-1?

GPT-1 was published by OpenAI, based in United States of America, categorised as industry.

When was GPT-1 released?

GPT-1 was published in June 2018. 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 GPT-1 used for?

GPT-1 works in Language, and is recorded as handling question answering, Text classification, Language modeling. These are the areas it was designed around; they describe intent rather than a hard boundary.

Where can I download GPT-1?

The weights for GPT-1 are published, though we do not hold a repository link for it. This site calculates hardware requirements rather than hosting model files.

How much compute was used to train GPT-1?

Around 1.8 × 10¹⁹ FLOP, on NVIDIA Quadro P600. 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 GPT-1 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 GPT-1 is rarely worth using. Every figure here assumes the whole model is on the card.

Would two GPUs run GPT-1 faster?

Capacity adds across cards; throughput does not. Since 818 of the cards we track already hold GPT-1 on their own, a second card is rarely the answer here.

Why does the quantisation differ between cards for GPT-1?

A larger card holds a more accurate copy. Across the cards that run GPT-1, 1 compression levels are used; the floor control above pins it to one.

How accurate are these GPT-1 speed estimates?

They are calculated from specifications rather than measured, and each carries a range — 17,376–46,335 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 GPT-1?

The smallest card in our catalogue that holds GPT-1 is the Tesla C1080, with 4 GB of memory. It runs the model at Q8_0 using about 0.8 GB, and produces roughly 315 tokens per second. 818 cards in total can run it.

How fast is GPT-1 on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 28,959 tokens per second; the slowest that still runs it manages considerably less. Reading speed is around ten tokens per second, and 818 of the cards that can run GPT-1 clear that.

How much VRAM does GPT-1 need?

About 0.8 GB at Q8_0 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.

Source

Original publication

Record last updated 1 January 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