TRIMELMext (7M) TPS calculator

Open weights Princeton University 7M parameters May 2022

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 · 5,266 tok/s

Fastest card

B200

484,034 tok/s · 180 GB

Which GPUs can run TRIMELMext (7M)?

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
484,034 tok/s 290,420–774,454 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	0.7 GB	Q8_0	Comfortable
484,034 tok/s 290,420–774,454 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	0.7 GB	Q8_0	Comfortable
386,513 tok/s 231,908–618,421 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	0.7 GB	Q8_0	Comfortable
386,513 tok/s 231,908–618,421 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	0.7 GB	Q8_0	Comfortable
309,116 tok/s 185,470–494,586 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	0.7 GB	Q8_0	Comfortable
295,866 tok/s 177,519–473,385 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	0.7 GB	Q8_0	Comfortable
295,866 tok/s 177,519–473,385 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	0.7 GB	Q8_0	Comfortable
283,160 tok/s 169,896–453,055 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	0.7 GB	Q8_0	Comfortable
251,304 tok/s 150,783–402,087 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	0.7 GB	Q8_0	Comfortable
251,304 tok/s 150,783–402,087 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	0.7 GB	Q8_0	Comfortable
251,304 tok/s 150,783–402,087 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	0.7 GB	Q8_0	Comfortable
238,387 tok/s 143,032–381,418 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
203,294 tok/s 121,976–325,271 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
203,294 tok/s 121,976–325,271 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	0.7 GB	Q8_0	Comfortable
203,294 tok/s 121,976–325,271 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
203,294 tok/s 121,976–325,271 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
203,294 tok/s 121,976–325,271 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
154,794 tok/s 92,876–247,670 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	0.7 GB	Q8_0	Comfortable
154,794 tok/s 92,876–247,670 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	0.7 GB	Q8_0	Comfortable
128,995 tok/s 77,397–206,392 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	0.7 GB	Q8_0	Comfortable
126,242 tok/s 75,745–201,987 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	0.7 GB	Q8_0	Comfortable
123,429 tok/s 74,057–197,486 · low confidence	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	0.7 GB	Q8_0	Comfortable
123,429 tok/s 74,057–197,486 · low confidence	A100X NVIDIA	80 GB	2,040 GB/s	Jun 2021	0.7 GB	Q8_0	Comfortable
123,429 tok/s 74,057–197,486 · low confidence	A800 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Aug 2022	0.7 GB	Q8_0	Comfortable
123,429 tok/s 74,057–197,486 · 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: Princeton University
Organisation type: Academia
Country: United States of America
Published: 25 May 2022
Authors: Zexuan Zhong, Tao Lei, Danqi Chen

What it does

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

Domain: Language
Task: Language modeling/generation, Translation

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: 7M
Training data: 4,915,200,000 tokens
Epochs: 47.72

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: 2.1 × 10¹⁷ FLOP

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
Citations: 152
Benchmark data: TRIMELMext (7M)

Sources

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

Reference: Training Language Models with Memory Augmentation
Last updated: 25 May 2026

The extremes

The ten fastest GPUs that run TRIMELMext (7M)

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 TRIMELMext (7M)

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 it takes to run this model

Minimum card

Tesla C1080

Memory needed

0.7 GB

Fastest

484,034 tok/s

TRIMELMext (7M) is small enough at 7M parameters that hardware is rarely the obstacle — 818 of the cards we track can run it, including cards several years old.

The smallest card that holds it is the Tesla C1080 with 4 GB, running it at Q8_0 and producing around 5,266 tokens per second.

At the other end, a B200 generates roughly 484,034 tokens per second on it, on the strength of 8,000 GB/s of memory bandwidth.

Background

TRIMELMext (7M) was published by Princeton University, in United States of America, in May 2022. It comes out of academia.

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

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

The median result is around 13,591.7 tokens per second; 818 cards produce text faster than most people read it.

It is a dense model, so every parameter is read for every token produced. That makes speed track memory bandwidth almost exactly — a card with twice the bandwidth generates roughly twice as fast.

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.

What went into building it

Producing it required around 2.1 × 10¹⁷ FLOP of arithmetic, which is a statement about the training budget rather than about inference.

It was trained on about 4,915,200,000 tokens of text.

Step by step

How to choose a GPU for TRIMELMext (7M)

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

Look at what TRIMELMext (7M) actually needs — around 0.7 GB at Q8_0. No amount of processing power compensates for a card that cannot hold it.
02

Match the context to your actual use

Set the context to what you will actually use. The cache grows with the conversation, and it is the usual reason TRIMELMext (7M) stops fitting a card that seemed fine.
03

Decide how much compression you will accept

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 TRIMELMext (7M) by squeezing it further than you would want.
04

Rank by throughput rather than spec sheet

Ranking by tokens per second for TRIMELMext (7M) follows memory bandwidth, not core counts, which is why the B200 tops it at 484,034 tok/s.
05

Check the fit verdict before buying

The fit column separates cards that just manage TRIMELMext (7M) from those with room to spare. Buy for the second if the context might grow.
06

Check the card from the other side

Following a card through to its own page shows every other model it can hold, which is the question that follows once TRIMELMext (7M) is settled.

Answers

TRIMELMext (7M) — common questions

Can I run TRIMELMext (7M) 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 90,151 tokens per second — a comfortable fit.

Can I run TRIMELMext (7M) 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 55,204 tokens per second — a comfortable fit.

Can I run TRIMELMext (7M) 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 68,370 tokens per second — a comfortable fit.

Can I run TRIMELMext (7M) 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 81,076 tokens per second — a comfortable fit.

Is TRIMELMext (7M) open source?

Its weights are published, so TRIMELMext (7M) 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 TRIMELMext (7M) have?

TRIMELMext (7M) has 7M 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 TRIMELMext (7M)?

TRIMELMext (7M) was published by Princeton University, based in United States of America, categorised as academia.

When was TRIMELMext (7M) released?

TRIMELMext (7M) was published in May 2022. 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 TRIMELMext (7M) used for?

TRIMELMext (7M) works in Language, and is recorded as handling language modeling/generation, Translation. 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 TRIMELMext (7M)?

The weights for TRIMELMext (7M) 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 TRIMELMext (7M)?

Around 2.1 × 10¹⁷ 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 TRIMELMext (7M) if it does not fit in my GPU?

It can be split between the card and system memory, but TRIMELMext (7M) generates painfully slowly that way. Nothing on this page assumes offloading.

Would two GPUs run TRIMELMext (7M) faster?

A second card roughly doubles the memory available but not the generation rate. With 818 cards already able to run TRIMELMext (7M) alone, the case for pairing is weak.

Why does the quantisation differ between cards for TRIMELMext (7M)?

Because capacity varies, so does how hard TRIMELMext (7M) has to be squeezed — 1 distinct levels appear in the table above. Set a minimum quality to compare at one.

How accurate are these TRIMELMext (7M) 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 290,420–774,454 tok/s on the B200 rather than a single number.

What GPU do I need to run TRIMELMext (7M)?

The smallest card in our catalogue that holds TRIMELMext (7M) is the Tesla C1080, with 4 GB of memory. It runs the model at Q8_0 using about 0.7 GB, and produces roughly 5,266 tokens per second. 818 cards in total can run it.

How fast is TRIMELMext (7M) on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 484,034 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 TRIMELMext (7M) clear that.

How much VRAM does TRIMELMext (7M) 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.

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