Transformer-XL (257M) TPS calculator

Open weights Carnegie Mellon University (CMU),Google Brain 257M parameters January 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 · 143 tok/s

Fastest card

B200

13,184 tok/s · 180 GB

Which GPUs can run Transformer-XL (257M)?

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
13,184 tok/s 7,910–21,094 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	1.0 GB	Q8_0	Comfortable
13,184 tok/s 7,910–21,094 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	1.0 GB	Q8_0	Comfortable
10,528 tok/s 6,317–16,844 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	1.0 GB	Q8_0	Comfortable
10,528 tok/s 6,317–16,844 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	1.0 GB	Q8_0	Comfortable
8,420 tok/s 5,052–13,471 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	1.0 GB	Q8_0	Comfortable
8,059 tok/s 4,835–12,894 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	1.0 GB	Q8_0	Comfortable
8,059 tok/s 4,835–12,894 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	1.0 GB	Q8_0	Comfortable
7,713 tok/s 4,628–12,340 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	1.0 GB	Q8_0	Comfortable
6,845 tok/s 4,107–10,952 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	1.0 GB	Q8_0	Comfortable
6,845 tok/s 4,107–10,952 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	1.0 GB	Q8_0	Comfortable
6,845 tok/s 4,107–10,952 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	1.0 GB	Q8_0	Comfortable
6,493 tok/s 3,896–10,389 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	1.0 GB	Q8_0	Comfortable
5,537 tok/s 3,322–8,860 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	1.0 GB	Q8_0	Comfortable
5,537 tok/s 3,322–8,860 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	1.0 GB	Q8_0	Comfortable
5,537 tok/s 3,322–8,860 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	1.0 GB	Q8_0	Comfortable
5,537 tok/s 3,322–8,860 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	1.0 GB	Q8_0	Comfortable
5,537 tok/s 3,322–8,860 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	1.0 GB	Q8_0	Comfortable
4,216 tok/s 2,530–6,746 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	1.0 GB	Q8_0	Comfortable
4,216 tok/s 2,530–6,746 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	1.0 GB	Q8_0	Comfortable
3,513 tok/s 2,108–5,622 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	1.0 GB	Q8_0	Comfortable
3,439 tok/s 2,063–5,502 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	1.0 GB	Q8_0	Comfortable
3,362 tok/s 2,017–5,379 · low confidence	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	1.0 GB	Q8_0	Comfortable
3,362 tok/s 2,017–5,379 · low confidence	A100X NVIDIA	80 GB	2,040 GB/s	Jun 2021	1.0 GB	Q8_0	Comfortable
3,362 tok/s 2,017–5,379 · low confidence	A800 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Aug 2022	1.0 GB	Q8_0	Comfortable
3,362 tok/s 2,017–5,379 · low confidence	H100 CNX NVIDIA	80 GB	2,040 GB/s	Mar 2023	1.0 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: Carnegie Mellon University (CMU),Google Brain
Organisation type: Academia,Industry
Country: United States of America
Published: 9 January 2019
Authors: Zihang Dai, Zhilin Yang, Yiming Yang, Jaime Carbonell, Quoc V. Le, Ruslan Salakhutdinov

What it does

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

Domain: Language
Task: Language modeling/generation
Numerical format: FP32

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: 257M
Training data: 103,000,000 tokens
Epochs: 1,908
Batch size: 49,152

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: 3.8 × 10²⁰ FLOP
How it was established: Operation counting,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: Google TPU v3
Chips used: 32
Power draw: 29.7 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: 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: Discretionary
Record confidence: Speculative
Citations: 4,320
Benchmark data: Transformer-XL Large

Sources

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

Reference: Transformer-XL: Attentive Language Models Beyond a Fixed-Length Context
Last updated: 25 May 2026

The extremes

The ten fastest GPUs that run Transformer-XL (257M)

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 Transformer-XL (257M)

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

1.0 GB

Fastest

13,184 tok/s

Transformer-XL (257M) is small enough at 257M 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 143 tokens per second.

A B200 is the fastest we calculate for it: about 13,184 tokens per second, from 8,000 GB/s of memory bandwidth.

About this model

Transformer-XL (257M) was published by Carnegie Mellon University (CMU),Google Brain, in United States of America, in January 2019. The organisation is categorised as academia,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.

How fast it runs, and why

Half the cards that hold it manage more than 370.2 tokens per second, and 818 exceed reading speed outright.

Every weight participates in every token here, so bandwidth is the whole story: the ranking below is effectively a ranking of memory throughput.

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.

What went into building it

Producing it required around 3.8 × 10²⁰ FLOP of arithmetic, on Google TPU v3, which is a statement about the training budget rather than about inference.

It was trained on about 103,000,000 tokens of text.

Its inclusion criterion is discretionary.

Step by step

How to choose a GPU for Transformer-XL (257M)

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 Transformer-XL (257M) — around 1.0 GB at Q8_0. Capacity is the gate — a card either holds it or it does not.
02

Set the context length you will work at

Set the context to what you will actually use. The cache grows with the conversation, and it is the usual reason Transformer-XL (257M) stops fitting a card that seemed fine.
03

Decide how much compression you will accept

The quantisation column varies by card, because a bigger card holds a more accurate copy of Transformer-XL (257M) — Q8_0 on the smallest card that fits. Set a floor to hold the comparison at one level.
04

Compare tokens per second, not specifications

Ranking by tokens per second for Transformer-XL (257M) follows memory bandwidth, not core counts, which is why the B200 tops it at 13,184 tok/s.
05

Look at the headroom, not just the fit

Tight means Transformer-XL (257M) loads and works, with no room to raise the context later. Comfortable means you can. The difference matters more than a few tokens per second.
06

See what else that card runs

Following a card through to its own page shows every other model it can hold, which is the question that follows once Transformer-XL (257M) is settled.

Answers

Transformer-XL (257M) — common questions

How fast is Transformer-XL (257M) on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 13,184 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 Transformer-XL (257M) clear that.

How much VRAM does Transformer-XL (257M) need?

About 1.0 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 Transformer-XL (257M) on a 8 GB GPU?

Yes. A CMP 170HX 8 GB with 8 GB runs it at Q8_0, using about 1.0 GB and generating roughly 2,455 tokens per second — a comfortable fit.

Can I run Transformer-XL (257M) on a 12 GB GPU?

Yes. A GeForce RTX 3080 Ti with 12 GB runs it at Q8_0, using about 1.0 GB and generating roughly 1,504 tokens per second — a comfortable fit.

Can I run Transformer-XL (257M) on a 16 GB GPU?

Yes. A Tesla V100 SXM2 16 GB with 16 GB runs it at Q8_0, using about 1.0 GB and generating roughly 1,862 tokens per second — a comfortable fit.

Can I run Transformer-XL (257M) on a 24 GB GPU?

Yes. A GeForce RTX 5090 D V2 with 24 GB runs it at Q8_0, using about 1.0 GB and generating roughly 2,208 tokens per second — a comfortable fit.

Is Transformer-XL (257M) open source?

Its weights are published, so Transformer-XL (257M) 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 Transformer-XL (257M) have?

Transformer-XL (257M) has 257M parameters. Transformer-XL Large, Table 1. 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 Transformer-XL (257M)?

Transformer-XL (257M) was published by Carnegie Mellon University (CMU),Google Brain, based in United States of America, categorised as academia,Industry.

When was Transformer-XL (257M) released?

Transformer-XL (257M) was published in January 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 Transformer-XL (257M) used for?

Transformer-XL (257M) works in Language, and is recorded as handling 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.

Where can I download Transformer-XL (257M)?

The weights for Transformer-XL (257M) 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 Transformer-XL (257M)?

Around 3.8 × 10²⁰ FLOP, on Google TPU v3. 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 Transformer-XL (257M) if it does not fit in my GPU?

Only by offloading, which is usually a false economy: the part in system memory drags the whole thing down. Our figures for Transformer-XL (257M) assume it is fully resident.

Would two GPUs run Transformer-XL (257M) faster?

Two cards buy memory rather than speed. That matters for Transformer-XL (257M) only if one card cannot hold it — 818 can, so a second adds little.

Why does the quantisation differ between cards for Transformer-XL (257M)?

A larger card holds a more accurate copy. Across the cards that run Transformer-XL (257M), 1 compression levels are used; the floor control above pins it to one.

How accurate are these Transformer-XL (257M) speed estimates?

These are estimates with real error bars. The fastest result here, 7,910–21,094 tok/s on the B200, could reasonably land anywhere in its published range depending on which runtime you use.

What GPU do I need to run Transformer-XL (257M)?

The smallest card in our catalogue that holds Transformer-XL (257M) is the Tesla C1080, with 4 GB of memory. It runs the model at Q8_0 using about 1.0 GB, and produces roughly 143 tokens per second. 818 cards in total can run it.

Source

Original publication

Record last updated 25 May 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