AWD-LSTM-MoS + dynamic evaluation (PTB, 2018) TPS calculator

Open weights Peking University,Microsoft Research Asia 24M parameters September 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 · 1,536 tok/s

Fastest card

B200

141,176 tok/s · 180 GB

Which GPUs can run AWD-LSTM-MoS + dynamic evaluation (PTB, 2018)?

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
141,176 tok/s 84,706–225,882 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	0.7 GB	Q8_0	Comfortable
141,176 tok/s 84,706–225,882 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	0.7 GB	Q8_0	Comfortable
112,733 tok/s 67,640–180,373 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	0.7 GB	Q8_0	Comfortable
112,733 tok/s 67,640–180,373 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	0.7 GB	Q8_0	Comfortable
90,159 tok/s 54,095–144,254 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	0.7 GB	Q8_0	Comfortable
86,294 tok/s 51,776–138,071 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	0.7 GB	Q8_0	Comfortable
86,294 tok/s 51,776–138,071 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	0.7 GB	Q8_0	Comfortable
82,588 tok/s 49,553–132,141 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	0.7 GB	Q8_0	Comfortable
73,297 tok/s 43,978–117,275 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	0.7 GB	Q8_0	Comfortable
73,297 tok/s 43,978–117,275 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	0.7 GB	Q8_0	Comfortable
73,297 tok/s 43,978–117,275 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	0.7 GB	Q8_0	Comfortable
69,529 tok/s 41,718–111,247 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
59,294 tok/s 35,576–94,871 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
59,294 tok/s 35,576–94,871 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	0.7 GB	Q8_0	Comfortable
59,294 tok/s 35,576–94,871 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
59,294 tok/s 35,576–94,871 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
59,294 tok/s 35,576–94,871 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
45,148 tok/s 27,089–72,237 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	0.7 GB	Q8_0	Comfortable
45,148 tok/s 27,089–72,237 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	0.7 GB	Q8_0	Comfortable
37,624 tok/s 22,574–60,198 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	0.7 GB	Q8_0	Comfortable
36,821 tok/s 22,092–58,913 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	0.7 GB	Q8_0	Comfortable
36,000 tok/s 21,600–57,600 · low confidence	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	0.7 GB	Q8_0	Comfortable
36,000 tok/s 21,600–57,600 · low confidence	A100X NVIDIA	80 GB	2,040 GB/s	Jun 2021	0.7 GB	Q8_0	Comfortable
36,000 tok/s 21,600–57,600 · low confidence	A800 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Aug 2022	0.7 GB	Q8_0	Comfortable
36,000 tok/s 21,600–57,600 · low confidence	H100 CNX NVIDIA	80 GB	2,040 GB/s	Mar 2023	0.7 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: Peking University,Microsoft Research Asia
Organisation type: Academia,Industry
Country: China
Published: 18 September 2018
Authors: Chengyue Gong, Di He, Xu Tan, Tao Qin, Liwei Wang, Tie-Yan Liu

What it does

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

Domain: Language
Task: Language modeling, Translation, Text classification

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: 24M
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: Open (non-commercial)

How it is classified

Labels the source dataset applies when tracking notable models, and how confident it is in the entry.

Record confidence: Confident
Benchmark data: AWD-LSTM-MoS + dynamic evaluation (PTB, 2018)

Sources

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

Reference: FRAGE: Frequency-Agnostic Word Representation
Last updated: 11 February 2026

The extremes

The ten fastest GPUs that run AWD-LSTM-MoS + dynamic evaluation (PTB, 2018)

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 AWD-LSTM-MoS + dynamic evaluation (PTB, 2018)

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

Hardware requirements in practice

Minimum card

Tesla C1080

Memory needed

0.7 GB

Fastest

141,176 tok/s

AWD-LSTM-MoS + dynamic evaluation (PTB, 2018) is small enough at 24M parameters that hardware is rarely the obstacle — 818 of the cards we track can run it, including cards several years old.

The entry point is the Tesla C1080: 4 GB of memory, Q8_0 compression, roughly 1,536 tokens per second.

The quickest result comes from a B200 at around 141,176 tokens per second — its 8,000 GB/s of bandwidth is what buys that.

What this model is

AWD-LSTM-MoS + dynamic evaluation (PTB, 2018) was published by Peking University,Microsoft Research Asia, in China, in September 2018. academia,Industry is the category the publisher falls under.

It works in Language, and is recorded as doing language modeling, Translation, Text classification.

The weights are published, so it can be downloaded and run on your own hardware indefinitely, offline, with no account attached.

What decides the speed

Across every card that can run it, the middle of the range is about 3,964.2 tokens per second, and 818 of them clear the ten tokens per second that roughly matches reading speed.

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

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 AWD-LSTM-MoS + dynamic evaluation (PTB, 2018)

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 AWD-LSTM-MoS + dynamic evaluation (PTB, 2018) — around 0.7 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 AWD-LSTM-MoS + dynamic evaluation (PTB, 2018).
03

Choose how far you will compress it

The quantisation column varies by card, because a bigger card holds a more accurate copy of AWD-LSTM-MoS + dynamic evaluation (PTB, 2018) — Q8_0 on the smallest card that fits. Set a floor to hold the comparison at one level.
04

Sort by speed

Ranking by tokens per second for AWD-LSTM-MoS + dynamic evaluation (PTB, 2018) follows memory bandwidth, not core counts, which is why the B200 tops it at 141,176 tok/s.
05

Look at the headroom, not just the fit

The fit column separates cards that just manage AWD-LSTM-MoS + dynamic evaluation (PTB, 2018) from those with room to spare. Buy for the second if the context might grow.
06

Open the card you have settled on

Each card page repeats this sweep for every model we hold. It answers what else the hardware is good for, beyond AWD-LSTM-MoS + dynamic evaluation (PTB, 2018).

Answers

AWD-LSTM-MoS + dynamic evaluation (PTB, 2018) — common questions

How accurate are these AWD-LSTM-MoS + dynamic evaluation (PTB, 2018) 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 84,706–225,882 tok/s on the B200 rather than a single number.

What GPU do I need to run AWD-LSTM-MoS + dynamic evaluation (PTB, 2018)?

The smallest card in our catalogue that holds AWD-LSTM-MoS + dynamic evaluation (PTB, 2018) is the Tesla C1080, with 4 GB of memory. It runs the model at Q8_0 using about 0.7 GB, and produces roughly 1,536 tokens per second. 818 cards in total can run it.

How fast is AWD-LSTM-MoS + dynamic evaluation (PTB, 2018) on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 141,176 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 AWD-LSTM-MoS + dynamic evaluation (PTB, 2018) clear that.

How much VRAM does AWD-LSTM-MoS + dynamic evaluation (PTB, 2018) need?

About 0.7 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 AWD-LSTM-MoS + dynamic evaluation (PTB, 2018) on a 8 GB GPU?

Yes. A CMP 170HX 8 GB with 8 GB runs it at Q8_0, using about 0.7 GB and generating roughly 26,294 tokens per second — a comfortable fit.

Can I run AWD-LSTM-MoS + dynamic evaluation (PTB, 2018) on a 12 GB GPU?

Yes. A GeForce RTX 3080 Ti with 12 GB runs it at Q8_0, using about 0.7 GB and generating roughly 16,101 tokens per second — a comfortable fit.

Can I run AWD-LSTM-MoS + dynamic evaluation (PTB, 2018) on a 16 GB GPU?

Yes. A Tesla V100 SXM2 16 GB with 16 GB runs it at Q8_0, using about 0.7 GB and generating roughly 19,941 tokens per second — a comfortable fit.

Can I run AWD-LSTM-MoS + dynamic evaluation (PTB, 2018) on a 24 GB GPU?

Yes. A GeForce RTX 5090 D V2 with 24 GB runs it at Q8_0, using about 0.7 GB and generating roughly 23,647 tokens per second — a comfortable fit.

Is AWD-LSTM-MoS + dynamic evaluation (PTB, 2018) open source?

Its weights are published, so AWD-LSTM-MoS + dynamic evaluation (PTB, 2018) 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 AWD-LSTM-MoS + dynamic evaluation (PTB, 2018) have?

AWD-LSTM-MoS + dynamic evaluation (PTB, 2018) has 24M parameters. 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 AWD-LSTM-MoS + dynamic evaluation (PTB, 2018)?

AWD-LSTM-MoS + dynamic evaluation (PTB, 2018) was published by Peking University,Microsoft Research Asia, based in China, categorised as academia,Industry.

When was AWD-LSTM-MoS + dynamic evaluation (PTB, 2018) released?

AWD-LSTM-MoS + dynamic evaluation (PTB, 2018) was published in September 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 AWD-LSTM-MoS + dynamic evaluation (PTB, 2018) used for?

AWD-LSTM-MoS + dynamic evaluation (PTB, 2018) works in Language, and is recorded as handling language modeling, Translation, Text classification. Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.

Where can I download AWD-LSTM-MoS + dynamic evaluation (PTB, 2018)?

The weights for AWD-LSTM-MoS + dynamic evaluation (PTB, 2018) are published, though we do not hold a repository link for it. This site calculates hardware requirements rather than hosting model files.

Can I run AWD-LSTM-MoS + dynamic evaluation (PTB, 2018) if it does not fit in my GPU?

It can be split between the card and system memory, but AWD-LSTM-MoS + dynamic evaluation (PTB, 2018) generates painfully slowly that way. Nothing on this page assumes offloading.

Would two GPUs run AWD-LSTM-MoS + dynamic evaluation (PTB, 2018) faster?

A second card roughly doubles the memory available but not the generation rate. With 818 cards already able to run AWD-LSTM-MoS + dynamic evaluation (PTB, 2018) alone, the case for pairing is weak.

Why does the quantisation differ between cards for AWD-LSTM-MoS + dynamic evaluation (PTB, 2018)?

A larger card holds a more accurate copy. Across the cards that run AWD-LSTM-MoS + dynamic evaluation (PTB, 2018), 1 compression levels are used; the floor control above pins it to one.

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