GPT-2 (124M) TPS calculator

Open weights OpenAI 124M parameters February 2019

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 · 297 tok/s

Fastest card

B200

27,324 tok/s · 180 GB

Which GPUs can run GPT-2 (124M)?

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
27,324 tok/s 16,395–43,719 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	0.8 GB	Q8_0	Comfortable
27,324 tok/s 16,395–43,719 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	0.8 GB	Q8_0	Comfortable
21,819 tok/s 13,092–34,911 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	0.8 GB	Q8_0	Comfortable
21,819 tok/s 13,092–34,911 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	0.8 GB	Q8_0	Comfortable
17,450 tok/s 10,470–27,920 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	0.8 GB	Q8_0	Comfortable
16,702 tok/s 10,021–26,723 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	0.8 GB	Q8_0	Comfortable
16,702 tok/s 10,021–26,723 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	0.8 GB	Q8_0	Comfortable
15,985 tok/s 9,591–25,576 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	0.8 GB	Q8_0	Comfortable
14,187 tok/s 8,512–22,698 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	0.8 GB	Q8_0	Comfortable
14,187 tok/s 8,512–22,698 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	0.8 GB	Q8_0	Comfortable
14,187 tok/s 8,512–22,698 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	0.8 GB	Q8_0	Comfortable
13,457 tok/s 8,074–21,532 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	0.8 GB	Q8_0	Comfortable
11,476 tok/s 6,886–18,362 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	0.8 GB	Q8_0	Comfortable
11,476 tok/s 6,886–18,362 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	0.8 GB	Q8_0	Comfortable
11,476 tok/s 6,886–18,362 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	0.8 GB	Q8_0	Comfortable
11,476 tok/s 6,886–18,362 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	0.8 GB	Q8_0	Comfortable
11,476 tok/s 6,886–18,362 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	0.8 GB	Q8_0	Comfortable
8,738 tok/s 5,243–13,981 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	0.8 GB	Q8_0	Comfortable
8,738 tok/s 5,243–13,981 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	0.8 GB	Q8_0	Comfortable
7,282 tok/s 4,369–11,651 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	0.8 GB	Q8_0	Comfortable
7,127 tok/s 4,276–11,403 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	0.8 GB	Q8_0	Comfortable
6,968 tok/s 4,181–11,148 · low confidence	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	0.8 GB	Q8_0	Comfortable
6,968 tok/s 4,181–11,148 · low confidence	A100X NVIDIA	80 GB	2,040 GB/s	Jun 2021	0.8 GB	Q8_0	Comfortable
6,968 tok/s 4,181–11,148 · low confidence	A800 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Aug 2022	0.8 GB	Q8_0	Comfortable
6,968 tok/s 4,181–11,148 · 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: 14 February 2019
Authors: Alec Radford, Jeffrey Wu, Rewon Child, David Luan, Dario Amodei, Ilya Sutskever

What it does

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

Domain: Language
Task: Language modeling/generation

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: 124M
Training data: 10,666,666,667 tokens
Epochs: 100

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: 7.9 × 10²⁰ FLOP
How it was established: Operation counting

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.

Why it is tracked: Highly cited
Record confidence: Speculative
Citations: 26,463
Benchmark data: GPT-2 (117M)

Sources

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

Reference: Language Models are Unsupervised Multitask Learners
Last updated: 11 February 2026

The extremes

The ten fastest GPUs that run GPT-2 (124M)

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-2 (124M)

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

Tesla C1080

Memory needed

0.8 GB

Fastest

27,324 tok/s

GPT-2 (124M) is small enough at 124M parameters that hardware is rarely the obstacle — 818 of the cards we track can run it, including cards several years old.

The least hardware that works is a Tesla C1080. Its 4 GB is enough at Q8_0 compression, giving roughly 297 tokens per second.

The quickest result comes from a B200 at around 27,324 tokens per second — its 8,000 GB/s of bandwidth is what buys that.

Where it came from

GPT-2 (124M) was published by OpenAI, in United States of America, in February 2019. It comes out of industry.

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

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 is about 767.3 tokens per second, and 818 of them clear the ten tokens per second that roughly matches reading speed.

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.

Its internal architecture is not on file, so memory is approximated from the parameter count and marked accordingly. Expect the real figure to differ, more so at long context.

How it was trained

The training run consumed about 7.9 × 10²⁰ FLOP. That figure describes the cost of creating it and has no bearing on how quickly it generates text.

Around 10,666,666,667 tokens went into training it.

It is tracked in the underlying dataset for one reason in particular: highly cited.

Step by step

How to choose a GPU for GPT-2 (124M)

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 GPT-2 (124M) — 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-2 (124M).
03

Set a quality floor

Each card runs the least-compressed copy it can hold — Q8_0 on the smallest card that fits. Setting a floor drops the cards that only manage GPT-2 (124M) by squeezing it further than you would want.
04

Compare tokens per second, not specifications

The speed ordering for GPT-2 (124M) is effectively an ordering by memory bandwidth, which is why the B200 tops it at 27,324 tok/s.
05

Read the fit column last

The fit column separates cards that just manage GPT-2 (124M) from those with room to spare. Buy for the second if the context might grow.
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-2 (124M) alone — a card is usually bought for more than one model.

Answers

GPT-2 (124M) — common questions

When was GPT-2 (124M) released?

GPT-2 (124M) was published in February 2019. 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-2 (124M) used for?

GPT-2 (124M) works in Language, and is recorded as handling language modeling/generation. Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.

Where can I download GPT-2 (124M)?

The weights for GPT-2 (124M) 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-2 (124M)?

Around 7.9 × 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 GPT-2 (124M) if it does not fit in my GPU?

It can be split between the card and system memory, but GPT-2 (124M) generates painfully slowly that way. Nothing on this page assumes offloading.

Would two GPUs run GPT-2 (124M) faster?

Capacity adds across cards; throughput does not. Since 818 of the cards we track already hold GPT-2 (124M) on their own, a second card is rarely the answer here.

Why does the quantisation differ between cards for GPT-2 (124M)?

A larger card holds a more accurate copy. Across the cards that run GPT-2 (124M), 1 compression levels are used; the floor control above pins it to one.

How accurate are these GPT-2 (124M) speed estimates?

Every figure is derived from memory bandwidth and model size, not benchmarked. That is why each is published as a range such as 16,395–43,719 tok/s on the B200 rather than a single number.

What GPU do I need to run GPT-2 (124M)?

The smallest card in our catalogue that holds GPT-2 (124M) is the Tesla C1080, with 4 GB of memory. It runs the model at Q8_0 using about 0.8 GB, and produces roughly 297 tokens per second. 818 cards in total can run it.

How fast is GPT-2 (124M) on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 27,324 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-2 (124M) clear that.

How much VRAM does GPT-2 (124M) 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.

Can I run GPT-2 (124M) 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 5,089 tokens per second — a comfortable fit.

Can I run GPT-2 (124M) 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 3,116 tokens per second — a comfortable fit.

Can I run GPT-2 (124M) 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 3,860 tokens per second — a comfortable fit.

Can I run GPT-2 (124M) 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 4,577 tokens per second — a comfortable fit.

Is GPT-2 (124M) open source?

Its weights are published, so GPT-2 (124M) 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-2 (124M) have?

GPT-2 (124M) has 124M parameters. Note that the initial paper release stated GPT-2 small had 117M parameters. The official github repo notes that this was due to an error: https://github.com/openai/gpt-2?tab=readme-ov-file. 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-2 (124M)?

GPT-2 (124M) was published by OpenAI, based in United States of America, categorised as industry.

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