MADLAD-400 10B TPS calculator

Open weights Google DeepMind,Google Research 10.7B parameters September 2023

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

509 cards that can run it

818 cards we hold specifications for

Smallest card that fits

Xeon Phi 5110P

8 GB · Q4_K_M · 19.0 tok/s

Fastest card

B200

317 tok/s · 180 GB

Which GPUs can run MADLAD-400 10B?

Set the inputs, read the answer

A longer conversation needs more memory, which can push this model off smaller cards.

Hides cards that would only fit the model by compressing it below this point.

509 cards match

Calculating
Needs Quantisation Fit
317 tok/s

190–507 · low confidence

B200 NVIDIA 180 GB 8,000 GB/s Jan 2024 12.2 GB Q8_0 Comfortable
317 tok/s

190–507 · low confidence

B300 NVIDIA 288 GB 8,000 GB/s Sep 2025 12.2 GB Q8_0 Comfortable
253 tok/s

152–405 · low confidence

Radeon Instinct MI350X AMD 288 GB 8,190 GB/s Jan 2025 12.2 GB Q8_0 Comfortable
253 tok/s

152–405 · low confidence

Radeon Instinct MI355X AMD 288 GB 8,190 GB/s Jan 2025 12.2 GB Q8_0 Comfortable
202 tok/s

121–324 · low confidence

Radeon Instinct MI300 AMD 128 GB 6,550 GB/s Jan 2023 12.2 GB Q8_0 Comfortable
194 tok/s

116–310 · low confidence

H200 NVL NVIDIA 141 GB 4,890 GB/s Nov 2024 12.2 GB Q8_0 Comfortable
194 tok/s

116–310 · low confidence

H200 SXM 141 GB NVIDIA 141 GB 4,890 GB/s Nov 2024 12.2 GB Q8_0 Comfortable
185 tok/s

111–296 · low confidence

Radeon Instinct MI325X AMD 256 GB 6,000 GB/s Oct 2024 12.2 GB Q8_0 Comfortable
164 tok/s

99–263 · low confidence

Radeon Instinct MI300A AMD 128 GB 5,325 GB/s Dec 2023 12.2 GB Q8_0 Comfortable
164 tok/s

99–263 · low confidence

Radeon Instinct MI300X AMD 192 GB 5,325 GB/s Dec 2023 12.2 GB Q8_0 Comfortable
164 tok/s

99–263 · low confidence

Radeon Instinct MI308X AMD 192 GB 5,325 GB/s Dec 2023 12.2 GB Q8_0 Comfortable
156 tok/s

94–250 · low confidence

H100 NVL 94 GB NVIDIA 94 GB 3,940 GB/s Mar 2023 12.2 GB Q8_0 Comfortable
136 tok/s

82–218 · low confidence

CMP 170HX 8 GB NVIDIA 8 GB 1,490 GB/s Sep 2021 7.2 GB Q4_K_M Tight
133 tok/s

80–213 · low confidence

H100 PCIe 96 GB NVIDIA 96 GB 3,360 GB/s Mar 2023 12.2 GB Q8_0 Comfortable
133 tok/s

80–213 · low confidence

H100 SXM5 80 GB NVIDIA 80 GB 3,360 GB/s Oct 2022 12.2 GB Q8_0 Comfortable
133 tok/s

80–213 · low confidence

H100 SXM5 94 GB NVIDIA 94 GB 3,360 GB/s Mar 2023 12.2 GB Q8_0 Comfortable
133 tok/s

80–213 · low confidence

H100 SXM5 96 GB NVIDIA 96 GB 3,360 GB/s Mar 2023 12.2 GB Q8_0 Comfortable
133 tok/s

80–213 · low confidence

H800 SXM5 NVIDIA 80 GB 3,360 GB/s Mar 2023 12.2 GB Q8_0 Comfortable
110 tok/s

66–176 · low confidence

CMP 170HX 10 GB NVIDIA 10 GB 1,560 GB/s Sep 2021 8.4 GB Q5_K_M Tight
101 tok/s

61–162 · low confidence

Radeon Instinct MI250 AMD 128 GB 3,280 GB/s Nov 2021 12.2 GB Q8_0 Comfortable
101 tok/s

61–162 · low confidence

Radeon Instinct MI250X AMD 128 GB 3,280 GB/s Nov 2021 12.2 GB Q8_0 Comfortable
84.4 tok/s

51–135 · low confidence

Data Center GPU Max 1550 Intel 128 GB 3,280 GB/s Jan 2023 12.2 GB Q8_0 Comfortable
82.6 tok/s

50–132 · low confidence

Data Center GPU Max Subsystem Intel 128 GB 3,210 GB/s Jan 2023 12.2 GB Q8_0 Comfortable
80.8 tok/s

48–129 · low confidence

A100 SXM4 80 GB NVIDIA 80 GB 2,040 GB/s Nov 2020 12.2 GB Q8_0 Comfortable
80.8 tok/s

48–129 · low confidence

A100X NVIDIA 80 GB 2,040 GB/s Jun 2021 12.2 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
Google DeepMind,Google Research
Organisation type
Industry,Industry
Country
United States of America
Published
9 September 2023
Authors
Sneha Kudugunta, Isaac Caswell, Biao Zhang, Xavier Garcia, Christopher A. Choquette-Choo, Katherine Lee, Derrick Xin, Aditya Kusupati, Romi Stella, Ankur Bapna, Orhan Firat

What it does

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

Domain
Language
Task
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
10.7B

10.7B from appendix A.8

Training data
250,000,000,000 tokens

MADLAD-400, dataset released with paper, is 3T tokens: 'We introduce MADLAD-400, a manually audited, general domain 3T token monolingual dataset based on CommonCrawl, spanning 419 languages.' However, model in question was trained on only 250B tokens: "We then train and release a 10.7B-parameter multilingual machine translation model on 250 billion tokens covering over 450 languages using publicly available data"

Epochs
1

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
1.6 × 10²² FLOP

6ND = 10.7B * 250B = 1.6e22 'MADLAD-400-10B-MT is a multilingual machine translation model based on the T5 architecture that was trained on 250 billion tokens covering over 450 languages using publicly available data. ' 10.7B params from appendix A.8

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)

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
229

Sources

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

Reference
MADLAD-400: A Multilingual And Document-Level Large Audited Dataset
Last updated
25 May 2026

The extremes

What the numbers mean

What it takes to run this model

Minimum card

Xeon Phi 5110P

Memory needed

7.2 GB

Fastest

317 tok/s

MADLAD-400 10B is small enough at 10.7B parameters that hardware is rarely the obstacle — 509 of the cards we track can run it, including cards several years old.

The smallest card that holds it is the Xeon Phi 5110P with 8 GB, running it at Q4_K_M and producing around 19.0 tokens per second.

A B200 is the fastest we calculate for it: about 317 tokens per second, from 8,000 GB/s of memory bandwidth.

Where it came from

MADLAD-400 10B was published by Google DeepMind,Google Research, in United States of America, in September 2023. It comes out of industry,Industry.

It works in Language, and is recorded as doing translation.

Because its weights were released, nothing about running it depends on a provider staying available — it is yours once downloaded.

Understanding the speeds

The median result is around 20.5 tokens per second; 461 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.

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

Training it took roughly 1.6 × 10²² FLOP of computation — a measure of what producing the model cost, not of how fast it answers.

Around 250,000,000,000 tokens went into training it.

Step by step

How to choose a GPU for MADLAD-400 10B

The table above has already assessed every card we hold specifications for against this model. Getting to your answer takes six steps.

  1. 01

    Start from the memory column

    Look at what MADLAD-400 10B actually needs — around 7.2 GB at Q4_K_M. No amount of processing power compensates for a card that cannot hold it.

  2. 02

    Match the context to your actual use

    Longer conversations cost memory on top of what the weights need. Move the slider to your real working length before trusting any row for MADLAD-400 10B.

  3. 03

    Choose how far you will compress it

    Each card runs the least-compressed copy it can hold — Q4_K_M on the smallest card that fits. Setting a floor drops the cards that only manage MADLAD-400 10B by squeezing it further than you would want.

  4. 04

    Sort by speed

    Sort by speed to see how cards rank for MADLAD-400 10B. It will not match a gaming ordering — generation is bound by memory bandwidth, which is why the B200 tops it at 317 tok/s.

  5. 05

    Read the fit column last

    Tight means MADLAD-400 10B 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.

  6. 06

    See what else that card runs

    Every card name links to its own page, which runs the same calculation across the whole model catalogue. Worth a look before buying for MADLAD-400 10B alone — a card is usually bought for more than one model.

Answers

MADLAD-400 10B — common questions

01

Why does the quantisation differ between cards for MADLAD-400 10B?

Each card is shown running the least-compressed copy it can hold, and MADLAD-400 10B appears at 4 different compression levels across the cards that fit it. Bigger cards get the more accurate version.

02

How accurate are these MADLAD-400 10B speed estimates?

These are estimates with real error bars. The fastest result here, 190–507 tok/s on the B200, could reasonably land anywhere in its published range depending on which runtime you use.

03

What GPU do I need to run MADLAD-400 10B?

The smallest card in our catalogue that holds MADLAD-400 10B is the Xeon Phi 5110P, with 8 GB of memory. It runs the model at Q4_K_M using about 7.2 GB, and produces roughly 19.0 tokens per second. 509 cards in total can run it.

04

How fast is MADLAD-400 10B on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 317 tokens per second; the slowest that still runs it manages considerably less. Reading speed is around ten tokens per second, and 461 of the cards that can run MADLAD-400 10B clear that.

05

How much VRAM does MADLAD-400 10B need?

About 7.2 GB at Q4_K_M 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.

06

Can I run MADLAD-400 10B on a 8 GB GPU?

Yes. A CMP 170HX 8 GB with 8 GB runs it at Q4_K_M, using about 7.2 GB and generating roughly 136 tokens per second — a tight fit.

07

Can I run MADLAD-400 10B on a 12 GB GPU?

Yes. A GeForce RTX 3080 Ti with 12 GB runs it at Q6_K, using about 9.7 GB and generating roughly 52.5 tokens per second — a tight fit.

08

Can I run MADLAD-400 10B on a 16 GB GPU?

Yes. A Tesla V100 SXM2 16 GB with 16 GB runs it at Q8_0, using about 12.2 GB and generating roughly 44.7 tokens per second — a tight fit.

09

Can I run MADLAD-400 10B on a 24 GB GPU?

Yes. A GeForce RTX 5090 D V2 with 24 GB runs it at Q8_0, using about 12.2 GB and generating roughly 53.0 tokens per second — a comfortable fit.

10

Is MADLAD-400 10B open source?

Its weights are published, so MADLAD-400 10B 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.

11

How many parameters does MADLAD-400 10B have?

MADLAD-400 10B has 10.7B parameters. 10.7B from appendix A.8. 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.

12

Who created MADLAD-400 10B?

MADLAD-400 10B was published by Google DeepMind,Google Research, based in United States of America, categorised as industry,Industry.

13

When was MADLAD-400 10B released?

MADLAD-400 10B was published in September 2023. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.

14

What is MADLAD-400 10B used for?

MADLAD-400 10B works in Language, and is recorded as handling 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.

15

Where can I download MADLAD-400 10B?

The weights for MADLAD-400 10B are published, though we do not hold a repository link for it. This site calculates hardware requirements rather than hosting model files.

16

How much compute was used to train MADLAD-400 10B?

Around 1.6 × 10²² 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.

17

Can I run MADLAD-400 10B 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 MADLAD-400 10B is rarely worth using — the nearest miss we calculate is short by 1.8 GB. Every figure here assumes the whole model is on the card.

18

Would two GPUs run MADLAD-400 10B faster?

Two cards buy memory rather than speed. That matters for MADLAD-400 10B only if one card cannot hold it — 509 can, so a second adds little.

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.