Grounding Dino L TPS calculator

Open weights Tsinghua University,International Digital Economy Academy,Hong Kong University of Science and Technology (HKUST),Chinese University of Hong Kong (CUHK),Microsoft Research,South China University of Technology 341M parameters July 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

818 cards that can run it

818 cards we hold specifications for

Smallest card that fits

Tesla C1080

4 GB · Q8_0 · 108 tok/s

Fastest card

B200

9,936 tok/s · 180 GB

Which GPUs can run Grounding Dino L?

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
9,936 tok/s 5,962–15,898 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	1.1 GB	Q8_0	Comfortable
9,936 tok/s 5,962–15,898 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	1.1 GB	Q8_0	Comfortable
7,934 tok/s 4,761–12,695 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	1.1 GB	Q8_0	Comfortable
7,934 tok/s 4,761–12,695 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	1.1 GB	Q8_0	Comfortable
6,345 tok/s 3,807–10,153 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	1.1 GB	Q8_0	Comfortable
6,073 tok/s 3,644–9,718 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	1.1 GB	Q8_0	Comfortable
6,073 tok/s 3,644–9,718 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	1.1 GB	Q8_0	Comfortable
5,813 tok/s 3,488–9,300 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	1.1 GB	Q8_0	Comfortable
5,159 tok/s 3,095–8,254 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	1.1 GB	Q8_0	Comfortable
5,159 tok/s 3,095–8,254 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	1.1 GB	Q8_0	Comfortable
5,159 tok/s 3,095–8,254 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	1.1 GB	Q8_0	Comfortable
4,894 tok/s 2,936–7,830 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	1.1 GB	Q8_0	Comfortable
4,173 tok/s 2,504–6,677 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	1.1 GB	Q8_0	Comfortable
4,173 tok/s 2,504–6,677 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	1.1 GB	Q8_0	Comfortable
4,173 tok/s 2,504–6,677 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	1.1 GB	Q8_0	Comfortable
4,173 tok/s 2,504–6,677 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	1.1 GB	Q8_0	Comfortable
4,173 tok/s 2,504–6,677 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	1.1 GB	Q8_0	Comfortable
3,178 tok/s 1,907–5,084 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	1.1 GB	Q8_0	Comfortable
3,178 tok/s 1,907–5,084 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	1.1 GB	Q8_0	Comfortable
2,648 tok/s 1,589–4,237 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	1.1 GB	Q8_0	Comfortable
2,591 tok/s 1,555–4,146 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	1.1 GB	Q8_0	Comfortable
2,534 tok/s 1,520–4,054 · low confidence	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	1.1 GB	Q8_0	Comfortable
2,534 tok/s 1,520–4,054 · low confidence	A100X NVIDIA	80 GB	2,040 GB/s	Jun 2021	1.1 GB	Q8_0	Comfortable
2,534 tok/s 1,520–4,054 · low confidence	A800 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Aug 2022	1.1 GB	Q8_0	Comfortable
2,534 tok/s 1,520–4,054 · 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: Tsinghua University,International Digital Economy Academy,Hong Kong University of Science and Technology (HKUST),Chinese University of Hong Kong (CUHK),Microsoft Research,South China University of Technology
Organisation type: Academia,Academia,Academia,Industry,Academia
Country: China, Hong Kong, United States of America
Published: 19 July 2024
Authors: Shilong Liu, Zhaoyang Zeng, Tianhe Ren, Feng Li, Hao Zhang, Jie Yang, Qing Jiang, Chunyuan Li, Jianwei Yang, Hang Su, Jun Zhu, Lei Zhang

What it does

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

Domain: Vision
Task: Object detection, Image captioning

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: 341M
Training data: tokens
Batch size: 64

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
Chips used: 64
Power draw: 50.5 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

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: 4,066

Sources

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

Reference: Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection
Last updated: 25 May 2026

The extremes

The ten fastest GPUs that run Grounding Dino L

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 Grounding Dino L

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

The hardware side

Minimum card

Tesla C1080

Memory needed

1.1 GB

Fastest

9,936 tok/s

Grounding Dino L is small enough at 341M parameters that hardware is rarely the obstacle — 818 of the cards we track can run it, including cards several years old.

At the low end, a Tesla C1080 handles it — 4 GB, at Q8_0, for about 108 tokens per second.

At the other end, a B200 generates roughly 9,936 tokens per second on it, on the strength of 8,000 GB/s of memory bandwidth.

What this model is

Grounding Dino L was published by Tsinghua University,International Digital Economy Academy,Hong Kong University of Science and Technology (HKUST),Chinese University of Hong Kong (CUHK),Microsoft Research,South China University of Technology, in China, in July 2024. It comes out of academia,Academia,Academia,Industry,Academia.

It works in Vision, and is recorded as doing object detection, Image captioning.

The weights being open is what puts this page in the calculator rather than only in the catalogue: it is a model you can actually hold.

What decides the speed

The median result is around 279.0 tokens per second; 818 cards produce text faster than most people read it.

Every weight participates in every token here, so bandwidth is the whole story: the ranking below is effectively a ranking of memory throughput.

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.

Step by step

How to choose a GPU for Grounding Dino L

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

The table lists every card that can hold Grounding Dino L — around 1.1 GB at Q8_0. That figure, not the card's headline performance, is what decides whether it runs.
02

Decide how long your conversations run

Set the context to what you will actually use. The cache grows with the conversation, and it is the usual reason Grounding Dino L stops fitting a card that seemed fine.
03

Decide how much compression you will accept

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

Compare tokens per second, not specifications

Ranking by tokens per second for Grounding Dino L follows memory bandwidth, not core counts, which is why the B200 tops it at 9,936 tok/s.
05

Look at the headroom, not just the fit

The fit column separates cards that just manage Grounding Dino L 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 Grounding Dino L alone — a card is usually bought for more than one model.

Answers

Grounding Dino L — common questions

How accurate are these Grounding Dino L speed estimates?

These are estimates with real error bars. The fastest result here, 5,962–15,898 tok/s on the B200, could reasonably land anywhere in its published range depending on which runtime you use.

What GPU do I need to run Grounding Dino L?

The smallest card in our catalogue that holds Grounding Dino L is the Tesla C1080, with 4 GB of memory. It runs the model at Q8_0 using about 1.1 GB, and produces roughly 108 tokens per second. 818 cards in total can run it.

How fast is Grounding Dino L on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 9,936 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 Grounding Dino L clear that.

How much VRAM does Grounding Dino L 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 Grounding Dino L 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 1,851 tokens per second — a comfortable fit.

Can I run Grounding Dino L 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 1,133 tokens per second — a comfortable fit.

Can I run Grounding Dino L 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 1,403 tokens per second — a comfortable fit.

Can I run Grounding Dino L 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 1,664 tokens per second — a comfortable fit.

Is Grounding Dino L open source?

Its weights are published, so Grounding Dino L 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 Grounding Dino L have?

Grounding Dino L has 341M parameters. Table 4 Swin-L [32] as an image backbone. 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 Grounding Dino L?

Grounding Dino L was published by Tsinghua University,International Digital Economy Academy,Hong Kong University of Science and Technology (HKUST),Chinese University of Hong Kong (CUHK),Microsoft Research,South China University of Technology, based in China, categorised as academia,Academia,Academia,Industry,Academia.

When was Grounding Dino L released?

Grounding Dino L was published in July 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 Grounding Dino L used for?

Grounding Dino L works in Vision, and is recorded as handling object detection, Image captioning. Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.

Where can I download Grounding Dino L?

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

Can I run Grounding Dino L 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. Our figures for Grounding Dino L assume it is fully resident.

Would two GPUs run Grounding Dino L faster?

A second card roughly doubles the memory available but not the generation rate. With 818 cards already able to run Grounding Dino L alone, the case for pairing is weak.

Why does the quantisation differ between cards for Grounding Dino L?

Because capacity varies, so does how hard Grounding Dino L has to be squeezed — 1 distinct levels appear in the table above. Set a minimum quality to compare at one.

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