LLaVA-OV-7B TPS calculator

Open weights ByteDance,Nanyang Technological University,Chinese University of Hong Kong (CUHK),Hong Kong University of Science and Technology (HKUST) 7.6B parameters August 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

589 cards that can run it

818 cards we hold specifications for

Smallest card that fits

Tesla K20c

5 GB · Q3_K_M · 26.6 tok/s

Fastest card

B200

446 tok/s · 180 GB

Which GPUs can run LLaVA-OV-7B?

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

589 cards match

Calculating

					Needs	Quantisation	Fit
446 tok/s 267–713 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	8.8 GB	Q8_0	Comfortable
446 tok/s 267–713 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	8.8 GB	Q8_0	Comfortable
356 tok/s 214–570 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	8.8 GB	Q8_0	Comfortable
356 tok/s 214–570 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	8.8 GB	Q8_0	Comfortable
285 tok/s 171–456 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	8.8 GB	Q8_0	Comfortable
273 tok/s 164–436 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	8.8 GB	Q8_0	Comfortable
273 tok/s 164–436 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	8.8 GB	Q8_0	Comfortable
261 tok/s 156–417 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	8.8 GB	Q8_0	Comfortable
231 tok/s 139–370 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	8.8 GB	Q8_0	Comfortable
231 tok/s 139–370 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	8.8 GB	Q8_0	Comfortable
231 tok/s 139–370 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	8.8 GB	Q8_0	Comfortable
220 tok/s 132–351 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	8.8 GB	Q8_0	Comfortable
187 tok/s 112–300 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	8.8 GB	Q8_0	Comfortable
187 tok/s 112–300 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	8.8 GB	Q8_0	Comfortable
187 tok/s 112–300 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	8.8 GB	Q8_0	Comfortable
187 tok/s 112–300 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	8.8 GB	Q8_0	Comfortable
187 tok/s 112–300 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	8.8 GB	Q8_0	Comfortable
143 tok/s 86–228 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	8.8 GB	Q8_0	Comfortable
143 tok/s 86–228 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	8.8 GB	Q8_0	Comfortable
121 tok/s 72–193 · low confidence	CMP 170HX 8 GB NVIDIA	8 GB	1,490 GB/s	Sep 2021	7.1 GB	Q6_K	Tight
119 tok/s 71–190 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	8.8 GB	Q8_0	Comfortable
116 tok/s 70–186 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	8.8 GB	Q8_0	Comfortable
114 tok/s 68–182 · low confidence	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	8.8 GB	Q8_0	Comfortable
114 tok/s 68–182 · low confidence	A100X NVIDIA	80 GB	2,040 GB/s	Jun 2021	8.8 GB	Q8_0	Comfortable
114 tok/s 68–182 · low confidence	A800 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Aug 2022	8.8 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: ByteDance,Nanyang Technological University,Chinese University of Hong Kong (CUHK),Hong Kong University of Science and Technology (HKUST)
Organisation type: Industry,Academia,Academia,Academia
Country: China, Singapore, Hong Kong
Published: 6 August 2024
Authors: Bo Li, Yuanhan Zhang, Dong Guo, Renrui Zhang, Feng Li, Hao Zhang, Kaichen Zhang, Peiyuan Zhang, Yanwei Li, Ziwei Liu, Chunyuan Li

What it does

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

Domain: Multimodal, Video, Vision, Language
Task: Image captioning, Visual question answering, Video description, Object recognition, Action recognition, Language modeling/generation
Base model: Qwen2-7B
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: 7.6B
Training data: 945,138,000 tokens
Epochs: 1
Batch size: 256

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: Open source
Hugging Face: lmms-lab

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: 2,428

Sources

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

Reference: LLaVA-OneVision: Easy Visual Task Transfer
Last updated: 25 May 2026

The extremes

The ten fastest GPUs that run LLaVA-OV-7B

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 LLaVA-OV-7B

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 K20c

Memory needed

4.4 GB

Fastest

446 tok/s

LLaVA-OV-7B is small enough at 7.6B parameters that hardware is rarely the obstacle — 589 of the cards we track can run it, including cards several years old.

At the low end, a Tesla K20c handles it — 5 GB, at Q3_K_M, for about 26.6 tokens per second.

Top of the range is the B200, at roughly 446 tokens per second thanks to 8,000 GB/s of bandwidth.

Where it came from

LLaVA-OV-7B was published by ByteDance,Nanyang Technological University,Chinese University of Hong Kong (CUHK),Hong Kong University of Science and Technology (HKUST), in China, in August 2024. industry,Academia,Academia,Academia is the category the publisher falls under.

It works in Multimodal, Video, Vision, Language, and is recorded as doing image captioning, Visual question answering, Video description, Object recognition, Action recognition, Language modeling/generation.

It is derived from Qwen2-7B rather than trained from scratch, which is the usual way a specialised model is produced.

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 lmms-lab organisation on Hugging Face.

Understanding the speeds

Half the cards that hold it manage more than 24.1 tokens per second, and 557 exceed reading speed outright.

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.

Training and provenance

It was trained on about 945,138,000 tokens of text.

Step by step

How to choose a GPU for LLaVA-OV-7B

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

Every card here has been checked against LLaVA-OV-7B — around 4.4 GB at Q3_K_M. Capacity is the gate — a card either holds it or it does not.
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 LLaVA-OV-7B stops fitting a card that seemed fine.
03

Set a quality floor

Each card runs the least-compressed copy it can hold — Q3_K_M on the smallest card that fits. Setting a floor drops the cards that only manage LLaVA-OV-7B by squeezing it further than you would want.
04

Compare tokens per second, not specifications

Sort by speed to see how cards rank for LLaVA-OV-7B. It will not match a gaming ordering — generation is bound by memory bandwidth, which is why the B200 tops it at 446 tok/s.
05

Read the fit column last

The fit column separates cards that just manage LLaVA-OV-7B from those with room to spare. Buy for the second if the context might grow.
06

Open the card you have settled on

Following a card through to its own page shows every other model it can hold, which is the question that follows once LLaVA-OV-7B is settled.

Answers

LLaVA-OV-7B — common questions

Can I run LLaVA-OV-7B on a 8 GB GPU?

Yes. A CMP 170HX 8 GB with 8 GB runs it at Q6_K, using about 7.1 GB and generating roughly 121 tokens per second — a tight fit.

Can I run LLaVA-OV-7B on a 12 GB GPU?

Yes. A GeForce RTX 3080 Ti with 12 GB runs it at Q8_0, using about 8.8 GB and generating roughly 50.9 tokens per second — a comfortable fit.

Can I run LLaVA-OV-7B on a 16 GB GPU?

Yes. A Tesla V100 SXM2 16 GB with 16 GB runs it at Q8_0, using about 8.8 GB and generating roughly 63.0 tokens per second — a comfortable fit.

Can I run LLaVA-OV-7B on a 24 GB GPU?

Yes. A GeForce RTX 5090 D V2 with 24 GB runs it at Q8_0, using about 8.8 GB and generating roughly 74.7 tokens per second — a comfortable fit.

Is LLaVA-OV-7B open source?

Its weights are published, so LLaVA-OV-7B 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 LLaVA-OV-7B have?

LLaVA-OV-7B has 7.6B parameters. 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 LLaVA-OV-7B?

LLaVA-OV-7B was published by ByteDance,Nanyang Technological University,Chinese University of Hong Kong (CUHK),Hong Kong University of Science and Technology (HKUST), based in China, categorised as industry,Academia,Academia,Academia.

When was LLaVA-OV-7B released?

LLaVA-OV-7B was published in August 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 LLaVA-OV-7B used for?

LLaVA-OV-7B works in Multimodal, Video, Vision, Language, and is recorded as handling image captioning, Visual question answering, Video description, Object recognition, Action recognition, Language modeling/generation. These are the areas it was designed around; they describe intent rather than a hard boundary.

Where can I download LLaVA-OV-7B?

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

Can I run LLaVA-OV-7B 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 LLaVA-OV-7B is rarely worth using — the nearest miss we calculate is short by 1.7 GB. Every figure here assumes the whole model is on the card.

Would two GPUs run LLaVA-OV-7B faster?

Capacity adds across cards; throughput does not. Since 589 of the cards we track already hold LLaVA-OV-7B on their own, a second card is rarely the answer here.

Why does the quantisation differ between cards for LLaVA-OV-7B?

A larger card holds a more accurate copy. Across the cards that run LLaVA-OV-7B, 4 compression levels are used; the floor control above pins it to one.

How accurate are these LLaVA-OV-7B speed estimates?

These are estimates with real error bars. The fastest result here, 267–713 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 LLaVA-OV-7B?

The smallest card in our catalogue that holds LLaVA-OV-7B is the Tesla K20c, with 5 GB of memory. It runs the model at Q3_K_M using about 4.4 GB, and produces roughly 26.6 tokens per second. 589 cards in total can run it.

How fast is LLaVA-OV-7B on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 446 tokens per second; the slowest that still runs it manages considerably less. Reading speed is around ten tokens per second, and 557 of the cards that can run LLaVA-OV-7B clear that.

How much VRAM does LLaVA-OV-7B need?

About 4.4 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.

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