Pleias-RAG-350m TPS calculator

Open weights PleIAs 350M parameters April 2025

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 · 105 tok/s

Fastest card

B200

9,681 tok/s · 180 GB

Which GPUs can run Pleias-RAG-350m?

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
9,681 tok/s 5,808–15,489 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	1.1 GB	Q8_0	Comfortable
9,681 tok/s 5,808–15,489 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	1.1 GB	Q8_0	Comfortable
7,730 tok/s 4,638–12,368 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	1.1 GB	Q8_0	Comfortable
7,730 tok/s 4,638–12,368 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	1.1 GB	Q8_0	Comfortable
6,182 tok/s 3,709–9,892 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	1.1 GB	Q8_0	Comfortable
5,917 tok/s 3,550–9,468 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	1.1 GB	Q8_0	Comfortable
5,917 tok/s 3,550–9,468 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	1.1 GB	Q8_0	Comfortable
5,663 tok/s 3,398–9,061 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	1.1 GB	Q8_0	Comfortable
5,026 tok/s 3,016–8,042 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	1.1 GB	Q8_0	Comfortable
5,026 tok/s 3,016–8,042 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	1.1 GB	Q8_0	Comfortable
5,026 tok/s 3,016–8,042 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	1.1 GB	Q8_0	Comfortable
4,768 tok/s 2,861–7,628 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	1.1 GB	Q8_0	Comfortable
4,066 tok/s 2,440–6,505 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	1.1 GB	Q8_0	Comfortable
4,066 tok/s 2,440–6,505 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	1.1 GB	Q8_0	Comfortable
4,066 tok/s 2,440–6,505 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	1.1 GB	Q8_0	Comfortable
4,066 tok/s 2,440–6,505 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	1.1 GB	Q8_0	Comfortable
4,066 tok/s 2,440–6,505 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	1.1 GB	Q8_0	Comfortable
3,096 tok/s 1,858–4,953 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	1.1 GB	Q8_0	Comfortable
3,096 tok/s 1,858–4,953 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	1.1 GB	Q8_0	Comfortable
2,580 tok/s 1,548–4,128 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	1.1 GB	Q8_0	Comfortable
2,525 tok/s 1,515–4,040 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	1.1 GB	Q8_0	Comfortable
2,469 tok/s 1,481–3,950 · low confidence	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	1.1 GB	Q8_0	Comfortable
2,469 tok/s 1,481–3,950 · low confidence	A100X NVIDIA	80 GB	2,040 GB/s	Jun 2021	1.1 GB	Q8_0	Comfortable
2,469 tok/s 1,481–3,950 · low confidence	A800 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Aug 2022	1.1 GB	Q8_0	Comfortable
2,469 tok/s 1,481–3,950 · low confidence	H100 CNX NVIDIA	80 GB	2,040 GB/s	Mar 2023	1.1 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: PleIAs
Organisation type: Industry
Country: France
Published: 25 April 2025
Authors: Pierre-Carl Langlais, Pavel Chizhov, Mattia Nee, Carlos Rosas Hinostroza, Matthieu Delsart, Irène Girard, Othman Hicheur, Anastasia Stasenko, Ivan P. Yamshchikov

What it does

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

Domain: Language
Task: Retrieval-augmented generation, Language modeling/generation, Question answering, Search, Text summarization, Translation
Base model: Pleias 1.0 350m

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: 350M
Training data: tokens
Epochs: 2

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: 2.7 × 10²¹ FLOP
How it was established: Operation counting
Fine-tuning compute: 4 × 10¹⁹ FLOP

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 H100 SXM5 80GB
Chips used: 16
Power draw: 22.0 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: Unreleased
Hugging Face: PleIAs

How it is classified

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

Record confidence: Confident

Sources

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

Reference: Even Small Reasoners Should Quote Their Sources: Introducing the Pleias-RAG Model Family
Last updated: 28 November 2025

The extremes

The ten fastest GPUs that run Pleias-RAG-350m

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 Pleias-RAG-350m

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

1.1 GB

Fastest

9,681 tok/s

Pleias-RAG-350m is small enough at 350M 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 105 tokens per second.

At the other end, a B200 generates roughly 9,681 tokens per second on it, on the strength of 8,000 GB/s of memory bandwidth.

Where it came from

Pleias-RAG-350m was published by PleIAs, in France, in April 2025. The organisation is categorised as industry.

It works in Language, and is recorded as doing retrieval-augmented generation, Language modeling/generation, Question answering, Search, Text summarization, Translation.

It builds on Pleias 1.0 350m, which is why it shares that model's general shape and size.

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. It is published under the PleIAs organisation on Hugging Face.

Understanding the speeds

The median result is around 271.8 tokens per second; 817 cards produce text faster than most people read it.

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.

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.

Training and provenance

Producing it required around 2.7 × 10²¹ FLOP of arithmetic, on NVIDIA H100 SXM5 80GB, which is a statement about the training budget rather than about inference.

Step by step

How to choose a GPU for Pleias-RAG-350m

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

The table lists every card that can hold Pleias-RAG-350m — around 1.1 GB at Q8_0. That figure, not the card's headline performance, is what decides whether it runs.
02

Set the context length you will work at

The conversation occupies memory too, and grows as it goes. Set the slider to the length you expect: at long context Pleias-RAG-350m can slip off a card that handles short questions easily.
03

Decide how much compression you will accept

The quantisation column varies by card, because a bigger card holds a more accurate copy of Pleias-RAG-350m — Q8_0 on the smallest card that fits. Set a floor to hold the comparison at one level.
04

Sort by speed

The speed ordering for Pleias-RAG-350m is effectively an ordering by memory bandwidth, which is why the B200 tops it at 9,681 tok/s.
05

Read the fit column last

The fit column separates cards that just manage Pleias-RAG-350m 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 Pleias-RAG-350m alone — a card is usually bought for more than one model.

Answers

Pleias-RAG-350m — common questions

Would two GPUs run Pleias-RAG-350m faster?

A second card roughly doubles the memory available but not the generation rate. With 818 cards already able to run Pleias-RAG-350m alone, the case for pairing is weak.

Why does the quantisation differ between cards for Pleias-RAG-350m?

A larger card holds a more accurate copy. Across the cards that run Pleias-RAG-350m, 1 compression levels are used; the floor control above pins it to one.

How accurate are these Pleias-RAG-350m 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 5,808–15,489 tok/s on the B200 rather than a single number.

What GPU do I need to run Pleias-RAG-350m?

The smallest card in our catalogue that holds Pleias-RAG-350m is the Tesla C1080, with 4 GB of memory. It runs the model at Q8_0 using about 1.1 GB, and produces roughly 105 tokens per second. 818 cards in total can run it.

How fast is Pleias-RAG-350m on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 9,681 tokens per second; the slowest that still runs it manages considerably less. Reading speed is around ten tokens per second, and 817 of the cards that can run Pleias-RAG-350m clear that.

How much VRAM does Pleias-RAG-350m need?

About 1.1 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 Pleias-RAG-350m on a 8 GB GPU?

Yes. A CMP 170HX 8 GB with 8 GB runs it at Q8_0, using about 1.1 GB and generating roughly 1,803 tokens per second — a comfortable fit.

Can I run Pleias-RAG-350m on a 12 GB GPU?

Yes. A GeForce RTX 3080 Ti with 12 GB runs it at Q8_0, using about 1.1 GB and generating roughly 1,104 tokens per second — a comfortable fit.

Can I run Pleias-RAG-350m on a 16 GB GPU?

Yes. A Tesla V100 SXM2 16 GB with 16 GB runs it at Q8_0, using about 1.1 GB and generating roughly 1,367 tokens per second — a comfortable fit.

Can I run Pleias-RAG-350m on a 24 GB GPU?

Yes. A GeForce RTX 5090 D V2 with 24 GB runs it at Q8_0, using about 1.1 GB and generating roughly 1,622 tokens per second — a comfortable fit.

Is Pleias-RAG-350m open source?

Its weights are published, so Pleias-RAG-350m 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 Pleias-RAG-350m have?

Pleias-RAG-350m has 350M parameters. 350M. 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 Pleias-RAG-350m?

Pleias-RAG-350m was published by PleIAs, based in France, categorised as industry.

When was Pleias-RAG-350m released?

Pleias-RAG-350m was published in April 2025.

What is Pleias-RAG-350m used for?

Pleias-RAG-350m works in Language, and is recorded as handling retrieval-augmented generation, Language modeling/generation, Question answering, Search, Text summarization, Translation. Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.

Where can I download Pleias-RAG-350m?

Its weights are published under the PleIAs organisation on Hugging Face. We do not host model files — this site calculates what hardware is needed to run them.

How much compute was used to train Pleias-RAG-350m?

Around 2.7 × 10²¹ FLOP, on NVIDIA H100 SXM5 80GB. 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 Pleias-RAG-350m 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 Pleias-RAG-350m is rarely worth using. Every figure here assumes the whole model is on the card.

Source

Original publication

Record last updated 28 November 2025

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