DBRX TPS calculator

Open weights Databricks 132B parameters March 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

38 of 818 cards that can run it

Smallest card that fits

RTX PRO 5000 72 GB Blackwell

72 GB · Q3_K_M · 42.5 tok/s

Fastest card

H100 NVL 94 GB

107 tok/s · 94 GB

Which GPUs can run DBRX?

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

38 cards match

Calculating

					Needs	Quantisation	Fit
107 tok/s 64–171 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	79.5 GB	Q4_K_M	Tight
97.1 tok/s 58–155 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	71.8 GB	IQ4_XS	Tight
97.1 tok/s 58–155 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	71.8 GB	IQ4_XS	Tight
94.1 tok/s 56–151 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	141.0 GB	Q8_0	Tight
94.1 tok/s 56–151 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	141.0 GB	Q8_0	Comfortable
91.3 tok/s 55–146 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	79.5 GB	Q4_K_M	Tight
91.3 tok/s 55–146 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	79.5 GB	Q4_K_M	Tight
91.3 tok/s 55–146 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	79.5 GB	Q4_K_M	Tight
87.3 tok/s 52–140 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	110.3 GB	Q6_K	Tight
83.6 tok/s 50–134 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	110.3 GB	Q6_K	Tight
83.6 tok/s 50–134 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	110.3 GB	Q6_K	Tight
75.2 tok/s 45–120 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	141.0 GB	Q8_0	Comfortable
75.2 tok/s 45–120 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	141.0 GB	Q8_0	Comfortable
71.0 tok/s 43–114 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	110.3 GB	Q6_K	Tight
58.9 tok/s 35–94 · low confidence	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	71.8 GB	IQ4_XS	Tight
58.9 tok/s 35–94 · low confidence	A100X NVIDIA	80 GB	2,040 GB/s	Jun 2021	71.8 GB	IQ4_XS	Tight
58.9 tok/s 35–94 · low confidence	A800 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Aug 2022	71.8 GB	IQ4_XS	Tight
58.9 tok/s 35–94 · low confidence	H100 CNX NVIDIA	80 GB	2,040 GB/s	Mar 2023	71.8 GB	IQ4_XS	Tight
58.9 tok/s 35–94 · low confidence	H100 PCIe 80 GB NVIDIA	80 GB	2,040 GB/s	Oct 2022	71.8 GB	IQ4_XS	Tight
58.9 tok/s 35–94 · low confidence	H800 PCIe 80 GB NVIDIA	80 GB	2,040 GB/s	Mar 2023	71.8 GB	IQ4_XS	Tight
56.0 tok/s 34–90 · low confidence	A100 PCIe 80 GB NVIDIA	80 GB	1,940 GB/s	Jun 2021	71.8 GB	IQ4_XS	Tight
56.0 tok/s 34–90 · low confidence	A800 PCIe 80 GB NVIDIA	80 GB	1,940 GB/s	Nov 2022	71.8 GB	IQ4_XS	Tight
55.1 tok/s 33–88 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	141.0 GB	Q8_0	Comfortable
48.9 tok/s 29–78 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	141.0 GB	Q8_0	Comfortable
48.9 tok/s 29–78 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	141.0 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: Databricks
Organisation type: Industry
Country: United States of America
Published: 27 March 2024
Authors: Mosaic Research Team

What it does

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

Domain: Language
Task: Chat, Code generation
Numerical format: BF16

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: 132B
Training data: 12,000,000,000,000 tokens
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: 2.6 × 10²⁴ FLOP
How it was established: Operation counting

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

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 (restricted use)
Training code: Unreleased
Hugging Face: databricks

How it is classified

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

Likely above 10²³ FLOP: Yes
Why it is tracked: Training cost
Record confidence: Confident

Sources

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

Reference: Introducing DBRX: A New State-of-the-Art Open LLM
Last updated: 28 November 2025

The extremes

The ten fastest GPUs for DBRX

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 DBRX

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

RTX PRO 5000 72 GB Blackwell

Memory needed

64.2 GB

Fastest

107 tok/s

DBRX sits at 132B parameters, which puts it above consumer hardware and into the range where a card is bought for this purpose rather than repurposed for it. 38 of the cards we track can hold it.

The smallest card that holds it is the RTX PRO 5000 72 GB Blackwell with 72 GB, running it at Q3_K_M and producing around 42.5 tokens per second.

At the other end, a H100 NVL 94 GB generates roughly 107 tokens per second on it, on the strength of 3,940 GB/s of memory bandwidth.

What this model is

DBRX was published by Databricks, in United States of America, in March 2024. industry is the category the publisher falls under.

It works in Language, and is recorded as doing chat, Code generation.

The weights are published, so it can be downloaded and run on your own hardware indefinitely, offline, with no account attached. It is published under the databricks organisation on Hugging Face.

What decides the speed

Across every card that can run it, the middle of the range is about 58.9 tokens per second, and 36 of them clear the ten tokens per second that roughly matches reading speed.

Mixture-of-experts routing is why the speeds here look high for the parameter count. Only a fraction is read per token, but the whole thing has to be loaded.

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.

How it was trained

The training run consumed about 2.6 × 10²⁴ FLOP, on NVIDIA H100 SXM5 80GB. That figure describes the cost of creating it and has no bearing on how quickly it generates text.

It was trained on about 12,000,000,000,000 tokens of text.

Its inclusion criterion is training cost.

Step by step

How to choose a GPU for DBRX

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

01

Start from the memory column

Look at what DBRX actually needs — around 64.2 GB at Q3_K_M. 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 DBRX stops fitting a card that seemed fine.
03

Decide how much compression you will accept

Compression is what makes DBRX fit smaller cards, at some cost in accuracy — Q3_K_M on the smallest card that fits. A minimum quality removes the ones that go too far.
04

Rank by throughput rather than spec sheet

Sort by speed to see how cards rank for DBRX. It will not match a gaming ordering — generation is bound by memory bandwidth, which is why the H100 NVL 94 GB tops it at 107 tok/s.
05

Read the fit column last

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

See what else that card runs

Each card page repeats this sweep for every model we hold. It answers what else the hardware is good for, beyond DBRX.

Answers

DBRX — common questions

How fast is DBRX on a GPU?

It depends on the card. The quickest we calculate is a H100 NVL 94 GB at about 107 tokens per second; the slowest that still runs it manages considerably less. Reading speed is around ten tokens per second, and 36 of the cards that can run DBRX clear that.

How much VRAM does DBRX need?

About 64.2 GB at Q3_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.

Is DBRX open source?

Its weights are published, so DBRX 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 DBRX have?

DBRX has 132B parameters. 132B mixture of experts. 36B parameters active per inference. 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 DBRX?

DBRX was published by Databricks, based in United States of America, categorised as industry.

When was DBRX released?

DBRX was published in March 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.

What is DBRX used for?

DBRX works in Language, and is recorded as handling chat, Code generation. 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 DBRX?

Its weights are published under the databricks 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 DBRX?

Around 2.6 × 10²⁴ 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 DBRX 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 21.9 GB. Our figures for DBRX assume it is fully resident.

Would two GPUs run DBRX faster?

Capacity adds across cards; throughput does not. Since 38 of the cards we track already hold DBRX on their own, a second card is rarely the answer here.

Why does the quantisation differ between cards for DBRX?

Each card is shown running the least-compressed copy it can hold, and DBRX appears at 5 different compression levels across the cards that fit it. Bigger cards get the more accurate version.

How accurate are these DBRX speed estimates?

They are calculated from specifications rather than measured, and each carries a range — 64–171 tok/s on the H100 NVL 94 GB, for instance. The same model and card vary by thirty to fifty per cent depending on the inference software and its version.

What GPU do I need to run DBRX?

The smallest card in our catalogue that holds DBRX is the RTX PRO 5000 72 GB Blackwell, with 72 GB of memory. It runs the model at Q3_K_M using about 64.2 GB, and produces roughly 42.5 tokens per second. 38 cards in total can run it.

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