Minitron 8B TPS calculator

Open weights NVIDIA 8.3B parameters November 2024

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

582 cards that can run it

818 cards we hold specifications for

Smallest card that fits

Quadro 6000

6 GB · IQ4_XS · 15.3 tok/s

Fastest card

B200

408 tok/s · 180 GB

Which GPUs can run Minitron 8B?

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.

582 cards match

Calculating
Needs Quantisation Fit
408 tok/s

245–653 · low confidence

B200 NVIDIA 180 GB 8,000 GB/s Jan 2024 9.6 GB Q8_0 Comfortable
408 tok/s

245–653 · low confidence

B300 NVIDIA 288 GB 8,000 GB/s Sep 2025 9.6 GB Q8_0 Comfortable
326 tok/s

196–522 · low confidence

Radeon Instinct MI350X AMD 288 GB 8,190 GB/s Jan 2025 9.6 GB Q8_0 Comfortable
326 tok/s

196–522 · low confidence

Radeon Instinct MI355X AMD 288 GB 8,190 GB/s Jan 2025 9.6 GB Q8_0 Comfortable
261 tok/s

156–417 · low confidence

Radeon Instinct MI300 AMD 128 GB 6,550 GB/s Jan 2023 9.6 GB Q8_0 Comfortable
250 tok/s

150–399 · low confidence

H200 NVL NVIDIA 141 GB 4,890 GB/s Nov 2024 9.6 GB Q8_0 Comfortable
250 tok/s

150–399 · low confidence

H200 SXM 141 GB NVIDIA 141 GB 4,890 GB/s Nov 2024 9.6 GB Q8_0 Comfortable
239 tok/s

143–382 · low confidence

Radeon Instinct MI325X AMD 256 GB 6,000 GB/s Oct 2024 9.6 GB Q8_0 Comfortable
212 tok/s

127–339 · low confidence

Radeon Instinct MI300A AMD 128 GB 5,325 GB/s Dec 2023 9.6 GB Q8_0 Comfortable
212 tok/s

127–339 · low confidence

Radeon Instinct MI300X AMD 192 GB 5,325 GB/s Dec 2023 9.6 GB Q8_0 Comfortable
212 tok/s

127–339 · low confidence

Radeon Instinct MI308X AMD 192 GB 5,325 GB/s Dec 2023 9.6 GB Q8_0 Comfortable
201 tok/s

121–322 · low confidence

H100 NVL 94 GB NVIDIA 94 GB 3,940 GB/s Mar 2023 9.6 GB Q8_0 Comfortable
171 tok/s

103–274 · low confidence

H100 PCIe 96 GB NVIDIA 96 GB 3,360 GB/s Mar 2023 9.6 GB Q8_0 Comfortable
171 tok/s

103–274 · low confidence

H100 SXM5 80 GB NVIDIA 80 GB 3,360 GB/s Oct 2022 9.6 GB Q8_0 Comfortable
171 tok/s

103–274 · low confidence

H100 SXM5 94 GB NVIDIA 94 GB 3,360 GB/s Mar 2023 9.6 GB Q8_0 Comfortable
171 tok/s

103–274 · low confidence

H100 SXM5 96 GB NVIDIA 96 GB 3,360 GB/s Mar 2023 9.6 GB Q8_0 Comfortable
171 tok/s

103–274 · low confidence

H800 SXM5 NVIDIA 80 GB 3,360 GB/s Mar 2023 9.6 GB Q8_0 Comfortable
136 tok/s

81–217 · low confidence

CMP 170HX 8 GB NVIDIA 8 GB 1,490 GB/s Sep 2021 6.7 GB Q5_K_M Tight
131 tok/s

78–209 · low confidence

Radeon Instinct MI250 AMD 128 GB 3,280 GB/s Nov 2021 9.6 GB Q8_0 Comfortable
131 tok/s

78–209 · low confidence

Radeon Instinct MI250X AMD 128 GB 3,280 GB/s Nov 2021 9.6 GB Q8_0 Comfortable
116 tok/s

69–185 · low confidence

CMP 170HX 10 GB NVIDIA 10 GB 1,560 GB/s Sep 2021 7.7 GB Q6_K Tight
109 tok/s

65–174 · low confidence

Data Center GPU Max 1550 Intel 128 GB 3,280 GB/s Jan 2023 9.6 GB Q8_0 Comfortable
106 tok/s

64–170 · low confidence

Data Center GPU Max Subsystem Intel 128 GB 3,210 GB/s Jan 2023 9.6 GB Q8_0 Comfortable
104 tok/s

62–167 · low confidence

A100 SXM4 80 GB NVIDIA 80 GB 2,040 GB/s Nov 2020 9.6 GB Q8_0 Comfortable
104 tok/s

62–167 · low confidence

A100X NVIDIA 80 GB 2,040 GB/s Jun 2021 9.6 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
NVIDIA
Organisation type
Industry
Country
United States of America
Published
4 November 2024
Authors
Saurav Muralidharan, Sharath Turuvekere Sreenivas, Raviraj Joshi, Marcin Chochowski, Mostofa Patwary, Mohammad Shoeybi, Bryan Catanzaro, Jan Kautz, Pavlo Molchanov

What it does

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

Domain
Language
Task
Language modeling/generation, Chat, Question answering
Base model
Nemotron-4 15B

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
8.3B

8.3B (table 2) (including 6.2B non-embedding parameters) Minitron-8B-Base is a large language model (LLM) obtained by pruning Nemotron-4 15B; specifically, we prune model embedding size, number of attention heads, and MLP intermediate dimension. Minitron-8B-Base uses a model embedding size of 4096, 48 attention heads, and an MLP intermediate dimension of 16384. It also uses Grouped-Query Attention (GQA) and Rotary Position Embeddings (RoPE).

Training data
tokens

94B training tokens (table 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.

How it was established
Operation counting
Fine-tuning compute
4.7 × 10²¹ FLOP

6*8300000000.00*94*10^9 = 4.6812e+21

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 A100 SXM4 80 GB
Chips used
8
Power draw
6.3 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 (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

Sources

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

Reference
Compact Language Models via Pruning and Knowledge Distillation
Last updated
28 November 2025

The extremes

What the numbers mean

Hardware requirements in practice

Minimum card

Quadro 6000

Memory needed

5.2 GB

Fastest

408 tok/s

Minitron 8B is small enough at 8.3B parameters that hardware is rarely the obstacle — 582 of the cards we track can run it, including cards several years old.

The entry point is the Quadro 6000: 6 GB of memory, IQ4_XS compression, roughly 15.3 tokens per second.

Top of the range is the B200, at roughly 408 tokens per second thanks to 8,000 GB/s of bandwidth.

Background

Minitron 8B was published by NVIDIA, in United States of America, in November 2024. The organisation is categorised as industry.

It works in Language, and is recorded as doing language modeling/generation, Chat, Question answering.

It is derived from Nemotron-4 15B rather than trained from scratch, which is the usual way a specialised model is produced.

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

Reading the throughput figures

Across every card that can run it, the middle of the range is about 22.9 tokens per second, and 551 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.

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.

Step by step

How to choose a GPU for Minitron 8B

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

  1. 01

    Check what it needs before anything else

    The table lists every card that can hold Minitron 8B — around 5.2 GB at IQ4_XS. That figure, not the card's headline performance, is what decides whether it runs.

  2. 02

    Set the context length you will work at

    Longer conversations cost memory on top of what the weights need. Move the slider to your real working length before trusting any row for Minitron 8B.

  3. 03

    Set a quality floor

    Each card runs the least-compressed copy it can hold — IQ4_XS on the smallest card that fits. Setting a floor drops the cards that only manage Minitron 8B by squeezing it further than you would want.

  4. 04

    Compare tokens per second, not specifications

    Sort by speed to see how cards rank for Minitron 8B. It will not match a gaming ordering — generation is bound by memory bandwidth, which is why the B200 tops it at 408 tok/s.

  5. 05

    Look at the headroom, not just the fit

    The fit column separates cards that just manage Minitron 8B from those with room to spare. Buy for the second if the context might grow.

  6. 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 Minitron 8B is settled.

Answers

Minitron 8B — common questions

01

How accurate are these Minitron 8B speed estimates?

They are calculated from specifications rather than measured, and each carries a range — 245–653 tok/s on the B200, for instance. The same model and card vary by thirty to fifty per cent depending on the inference software and its version.

02

What GPU do I need to run Minitron 8B?

The smallest card in our catalogue that holds Minitron 8B is the Quadro 6000, with 6 GB of memory. It runs the model at IQ4_XS using about 5.2 GB, and produces roughly 15.3 tokens per second. 582 cards in total can run it.

03

How fast is Minitron 8B on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 408 tokens per second; the slowest that still runs it manages considerably less. Reading speed is around ten tokens per second, and 551 of the cards that can run Minitron 8B clear that.

04

How much VRAM does Minitron 8B need?

About 5.2 GB at IQ4_XS 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.

05

Can I run Minitron 8B on a 8 GB GPU?

Yes. A CMP 170HX 8 GB with 8 GB runs it at Q5_K_M, using about 6.7 GB and generating roughly 136 tokens per second — a tight fit.

06

Can I run Minitron 8B on a 12 GB GPU?

Yes. A GeForce RTX 3080 Ti with 12 GB runs it at Q8_0, using about 9.6 GB and generating roughly 46.6 tokens per second — a tight fit.

07

Can I run Minitron 8B on a 16 GB GPU?

Yes. A Tesla V100 SXM2 16 GB with 16 GB runs it at Q8_0, using about 9.6 GB and generating roughly 57.7 tokens per second — a comfortable fit.

08

Can I run Minitron 8B on a 24 GB GPU?

Yes. A GeForce RTX 5090 D V2 with 24 GB runs it at Q8_0, using about 9.6 GB and generating roughly 68.4 tokens per second — a comfortable fit.

09

Is Minitron 8B open source?

Its weights are published, so Minitron 8B 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.

10

How many parameters does Minitron 8B have?

Minitron 8B has 8.3B parameters. 8.3B (table 2) (including 6.2B non-embedding parameters) Minitron-8B-Base is a large language model (LLM) obtained by pruning Nemotron-4 15B; specifically, we prune model embedding size, number of attention heads, and MLP intermediate dimension. Minitron-8B-Base uses a model embedding size of 4096, 48 attention heads, and an MLP intermediate dimension of 16384. It also uses Grouped-Query Attention (GQA) and Rotary Position Embeddings (RoPE). 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.

11

Who created Minitron 8B?

Minitron 8B was published by NVIDIA, based in United States of America, categorised as industry.

12

When was Minitron 8B released?

Minitron 8B was published in November 2024. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.

13

What is Minitron 8B used for?

Minitron 8B works in Language, and is recorded as handling language modeling/generation, Chat, Question answering. These are the areas it was designed around; they describe intent rather than a hard boundary.

14

Where can I download Minitron 8B?

The weights for Minitron 8B are published, though we do not hold a repository link for it. This site calculates hardware requirements rather than hosting model files.

15

Can I run Minitron 8B 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 — the nearest miss we calculate is short by 1.2 GB. Our figures for Minitron 8B assume it is fully resident.

16

Would two GPUs run Minitron 8B faster?

Two cards buy memory rather than speed. That matters for Minitron 8B only if one card cannot hold it — 582 can, so a second adds little.

17

Why does the quantisation differ between cards for Minitron 8B?

Because capacity varies, so does how hard Minitron 8B has to be squeezed — 4 distinct levels appear in the table above. Set a minimum quality to compare at one.

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.