InstructBLIP TPS calculator

Open weights Salesforce Research,Hong Kong University of Science and Technology (HKUST),Nanyang Technological University 13B parameters May 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 · Q3_K_M · 18.3 tok/s

Fastest card

B200

261 tok/s · 180 GB

Which GPUs can run InstructBLIP?

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

509 cards match

Calculating

					Needs	Quantisation	Fit
261 tok/s 156–417 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	14.6 GB	Q8_0	Comfortable
261 tok/s 156–417 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	14.6 GB	Q8_0	Comfortable
208 tok/s 125–333 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	14.6 GB	Q8_0	Comfortable
208 tok/s 125–333 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	14.6 GB	Q8_0	Comfortable
166 tok/s 100–266 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	14.6 GB	Q8_0	Comfortable
159 tok/s 96–255 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	14.6 GB	Q8_0	Comfortable
159 tok/s 96–255 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	14.6 GB	Q8_0	Comfortable
152 tok/s 91–244 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	14.6 GB	Q8_0	Comfortable
135 tok/s 81–217 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	14.6 GB	Q8_0	Comfortable
135 tok/s 81–217 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	14.6 GB	Q8_0	Comfortable
135 tok/s 81–217 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	14.6 GB	Q8_0	Comfortable
131 tok/s 79–210 · low confidence	CMP 170HX 8 GB NVIDIA	8 GB	1,490 GB/s	Sep 2021	7.1 GB	Q3_K_M	Tight
128 tok/s 77–205 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	14.6 GB	Q8_0	Comfortable
117 tok/s 70–188 · low confidence	CMP 170HX 10 GB NVIDIA	10 GB	1,560 GB/s	Sep 2021	8.6 GB	Q4_K_M	Tight
109 tok/s 66–175 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	14.6 GB	Q8_0	Comfortable
109 tok/s 66–175 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	14.6 GB	Q8_0	Comfortable
109 tok/s 66–175 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	14.6 GB	Q8_0	Comfortable
109 tok/s 66–175 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	14.6 GB	Q8_0	Comfortable
109 tok/s 66–175 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	14.6 GB	Q8_0	Comfortable
83.4 tok/s 50–133 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	14.6 GB	Q8_0	Comfortable
83.4 tok/s 50–133 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	14.6 GB	Q8_0	Comfortable
69.5 tok/s 42–111 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	14.6 GB	Q8_0	Comfortable
68.0 tok/s 41–109 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	14.6 GB	Q8_0	Comfortable
67.5 tok/s 41–108 · low confidence	RTX A5000-8Q NVIDIA	8 GB	768 GB/s	Apr 2021	7.1 GB	Q3_K_M	Tight
66.5 tok/s 40–106 · low confidence	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	14.6 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: Salesforce Research,Hong Kong University of Science and Technology (HKUST),Nanyang Technological University
Organisation type: Industry,Academia,Academia
Country: United States of America, Hong Kong, Singapore
Published: 11 May 2023
Authors: Wenliang Dai, Junnan Li, Dongxu Li, Anthony Meng Huat Tiong, Junqi Zhao, Weisheng Wang, Boyang Li, Pascale Fung, Steven Hoi

What it does

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

Domain: Multimodal, Language, Vision
Task: Visual question answering, Chat
Base model: Vicuna-13B v0
Numerical format: FP16

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: 13B
Training data: tokens

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: 1.9 × 10²⁰ FLOP
How it was established: Hardware
Fine-tuning compute: 1.9 × 10²⁰ FLOP

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 40 GB
Chips used: 16
Chip-hours: 576
Wall-clock time: 36 hours
Power draw: 12.7 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)
Training code: Open (non-commercial)
Hugging Face: Salesforce

How it is classified

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

Why it is tracked: SOTA improvement
Record confidence: Confident
Citations: 3,418

Sources

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

Reference: InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuning
Last updated: 25 June 2026

The extremes

The ten fastest GPUs that run InstructBLIP

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 InstructBLIP

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

Hardware requirements in practice

Minimum card

Xeon Phi 5110P

Memory needed

7.1 GB

Fastest

261 tok/s

InstructBLIP is small enough at 13B parameters that hardware is rarely the obstacle — 509 of the cards we track can run it, including cards several years old.

The entry point is the Xeon Phi 5110P: 8 GB of memory, Q3_K_M compression, roughly 18.3 tokens per second.

A B200 is the fastest we calculate for it: about 261 tokens per second, from 8,000 GB/s of memory bandwidth.

What this model is

InstructBLIP was published by Salesforce Research,Hong Kong University of Science and Technology (HKUST),Nanyang Technological University, in United States of America, in May 2023. industry,Academia,Academia is the category the publisher falls under.

It works in Multimodal, Language, Vision, and is recorded as doing visual question answering, Chat.

Its starting point was Vicuna-13B v0 — most models at this scale are adapted from an existing base rather than built from nothing.

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. It is published under the Salesforce organisation on Hugging Face.

What decides the speed

The median result is around 21.2 tokens per second; 459 cards produce text faster than most people read it.

It is a dense model, so every parameter is read for every token produced. That makes speed track memory bandwidth almost exactly — a card with twice the bandwidth generates roughly twice as fast.

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

The training run consumed about 1.9 × 10²⁰ FLOP, on NVIDIA A100 SXM4 40 GB. That figure describes the cost of creating it and has no bearing on how quickly it generates text.

The reason it appears in this catalogue at all is sOTA improvement.

Step by step

How to choose a GPU for InstructBLIP

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 InstructBLIP — around 7.1 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

The conversation occupies memory too, and grows as it goes. Set the slider to the length you expect: at long context InstructBLIP can slip off a card that handles short questions easily.
03

Set a quality floor

Compression is what makes InstructBLIP 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

The speed ordering for InstructBLIP is effectively an ordering by memory bandwidth, which is why the B200 tops it at 261 tok/s.
05

Check the fit verdict before buying

Tight means InstructBLIP 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.
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 InstructBLIP is settled.

Answers

InstructBLIP — common questions

When was InstructBLIP released?

InstructBLIP was published in May 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.

What is InstructBLIP used for?

InstructBLIP works in Multimodal, Language, Vision, and is recorded as handling visual question answering, Chat. Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.

Where can I download InstructBLIP?

Its weights are published under the Salesforce 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 InstructBLIP?

Around 1.9 × 10²⁰ FLOP, on NVIDIA A100 SXM4 40 GB. 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 InstructBLIP if it does not fit in my GPU?

It can be split between the card and system memory, but InstructBLIP generates painfully slowly that way — the nearest miss we calculate is short by 3.2 GB. Nothing on this page assumes offloading.

Would two GPUs run InstructBLIP faster?

A second card roughly doubles the memory available but not the generation rate. With 509 cards already able to run InstructBLIP alone, the case for pairing is weak.

Why does the quantisation differ between cards for InstructBLIP?

Because capacity varies, so does how hard InstructBLIP has to be squeezed — 5 distinct levels appear in the table above. Set a minimum quality to compare at one.

How accurate are these InstructBLIP speed estimates?

They are calculated from specifications rather than measured, and each carries a range — 156–417 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.

What GPU do I need to run InstructBLIP?

The smallest card in our catalogue that holds InstructBLIP is the Xeon Phi 5110P, with 8 GB of memory. It runs the model at Q3_K_M using about 7.1 GB, and produces roughly 18.3 tokens per second. 509 cards in total can run it.

How fast is InstructBLIP on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 261 tokens per second; the slowest that still runs it manages considerably less. Reading speed is around ten tokens per second, and 459 of the cards that can run InstructBLIP clear that.

How much VRAM does InstructBLIP need?

About 7.1 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.

Can I run InstructBLIP on a 8 GB GPU?

Yes. A CMP 170HX 8 GB with 8 GB runs it at Q3_K_M, using about 7.1 GB and generating roughly 131 tokens per second — a tight fit.

Can I run InstructBLIP on a 12 GB GPU?

Yes. A GeForce RTX 3080 Ti with 12 GB runs it at Q5_K_M, using about 10.1 GB and generating roughly 53.1 tokens per second — a tight fit.

Can I run InstructBLIP on a 16 GB GPU?

Yes. A Tesla V100 SXM2 16 GB with 16 GB runs it at Q6_K, using about 11.6 GB and generating roughly 53.5 tokens per second — a comfortable fit.

Can I run InstructBLIP on a 24 GB GPU?

Yes. A GeForce RTX 5090 D V2 with 24 GB runs it at Q8_0, using about 14.6 GB and generating roughly 43.7 tokens per second — a comfortable fit.

Is InstructBLIP open source?

Its weights are published, so InstructBLIP 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 InstructBLIP have?

InstructBLIP has 13B parameters. 13B form 2.6. 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 InstructBLIP?

InstructBLIP was published by Salesforce Research,Hong Kong University of Science and Technology (HKUST),Nanyang Technological University, based in United States of America, categorised as industry,Academia,Academia.

Source

Original publication

Record last updated 25 June 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