NVILA 15B TPS calculator

Open weights NVIDIA,Massachusetts Institute of Technology (MIT),University of California (UC) Berkeley,University of California San Diego,University of Washington,Tsinghua University 15B parameters December 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

306 cards that can run it

818 cards we hold specifications for

Smallest card that fits

P102-101

10 GB · IQ4_XS · 18.9 tok/s

Fastest card

B200

226 tok/s · 180 GB

Which GPUs can run NVILA 15B?

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

306 cards match

Calculating

					Needs	Quantisation	Fit
226 tok/s 136–361 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	16.8 GB	Q8_0	Comfortable
226 tok/s 136–361 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	16.8 GB	Q8_0	Comfortable
180 tok/s 108–289 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	16.8 GB	Q8_0	Comfortable
180 tok/s 108–289 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	16.8 GB	Q8_0	Comfortable
144 tok/s 87–231 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	16.8 GB	Q8_0	Comfortable
138 tok/s 83–221 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	16.8 GB	Q8_0	Comfortable
138 tok/s 83–221 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	16.8 GB	Q8_0	Comfortable
132 tok/s 79–211 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	16.8 GB	Q8_0	Comfortable
117 tok/s 70–188 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	16.8 GB	Q8_0	Comfortable
117 tok/s 70–188 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	16.8 GB	Q8_0	Comfortable
117 tok/s 70–188 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	16.8 GB	Q8_0	Comfortable
111 tok/s 67–178 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	16.8 GB	Q8_0	Comfortable
108 tok/s 65–173 · low confidence	CMP 170HX 10 GB NVIDIA	10 GB	1,560 GB/s	Sep 2021	8.9 GB	IQ4_XS	Tight
94.9 tok/s 57–152 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	16.8 GB	Q8_0	Comfortable
94.9 tok/s 57–152 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	16.8 GB	Q8_0	Comfortable
94.9 tok/s 57–152 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	16.8 GB	Q8_0	Comfortable
94.9 tok/s 57–152 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	16.8 GB	Q8_0	Comfortable
94.9 tok/s 57–152 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	16.8 GB	Q8_0	Comfortable
72.2 tok/s 43–116 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	16.8 GB	Q8_0	Comfortable
72.2 tok/s 43–116 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	16.8 GB	Q8_0	Comfortable
60.2 tok/s 36–96 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	16.8 GB	Q8_0	Comfortable
59.5 tok/s 36–95 · low confidence	GeForce RTX 3080 12 GB NVIDIA	12 GB	912 GB/s	Jan 2022	9.8 GB	Q4_K_M	Tight
59.5 tok/s 36–95 · low confidence	GeForce RTX 3080 Ti NVIDIA	12 GB	912 GB/s	May 2021	9.8 GB	Q4_K_M	Tight
58.9 tok/s 35–94 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	16.8 GB	Q8_0	Comfortable
57.6 tok/s 35–92 · low confidence	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	16.8 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: NVIDIA,Massachusetts Institute of Technology (MIT),University of California (UC) Berkeley,University of California San Diego,University of Washington,Tsinghua University
Organisation type: Industry,Academia,Academia,Academia,Academia,Academia
Country: United States of America, China
Published: 5 December 2024
Authors: Zhijian Liu, Ligeng Zhu, Baifeng Shi, Zhuoyang Zhang, Yuming Lou, Shang Yang, Haocheng Xi, Shiyi Cao, Yuxian Gu, Dacheng Li, Xiuyu Li, Yunhao Fang, Yukang Chen, Cheng-Yu Hsieh, De-An Huang, An-Chieh Cheng, Vishwesh Nath, Jinyi Hu, Sifei Liu, Ranjay Krishna, Daguang Xu, Xiaolong Wang, Pavlo Molchanov, Jan Kautz, Hongxu Yin, Song Han, Yao Lu

What it does

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

Domain: Vision, Language, Multimodal, Video
Task: Visual question answering, Video description, Language modeling/generation, Question answering, Character recognition (OCR)
Numerical format: FP8

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: 15B
Training data: tokens

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
Chips used: 128
Power draw: 176.2 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)
Hugging Face: Efficient-Large-Model

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

Sources

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

Reference: NVILA: Efficient Frontier Visual Language Models
Last updated: 28 November 2025

The extremes

The ten fastest GPUs that run NVILA 15B

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 NVILA 15B

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

P102-101

Memory needed

8.9 GB

Fastest

226 tok/s

NVILA 15B is small enough at 15B parameters that hardware is rarely the obstacle — 306 of the cards we track can run it, including cards several years old.

The entry point is the P102-101: 10 GB of memory, IQ4_XS compression, roughly 18.9 tokens per second.

At the other end, a B200 generates roughly 226 tokens per second on it, on the strength of 8,000 GB/s of memory bandwidth.

About this model

NVILA 15B was published by NVIDIA,Massachusetts Institute of Technology (MIT),University of California (UC) Berkeley,University of California San Diego,University of Washington,Tsinghua University, in United States of America, in December 2024. It comes out of industry,Academia,Academia,Academia,Academia,Academia.

It works in Vision, Language, Multimodal, Video, and is recorded as doing visual question answering, Video description, Language modeling/generation, Question answering, Character recognition (OCR).

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 Efficient-Large-Model organisation on Hugging Face.

How fast it runs, and why

Across every card that can run it, the middle of the range is about 21.1 tokens per second, and 266 of them clear the ten tokens per second that roughly matches reading speed.

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.

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.

Training and provenance

Its inclusion criterion is sOTA improvement.

Step by step

How to choose a GPU for NVILA 15B

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 NVILA 15B actually needs — around 8.9 GB at IQ4_XS. No amount of processing power compensates for a card that cannot hold it.
02

Match the context to your actual use

Longer conversations cost memory on top of what the weights need. Move the slider to your real working length before trusting any row for NVILA 15B.
03

Decide how much compression you will accept

The quantisation column varies by card, because a bigger card holds a more accurate copy of NVILA 15B — IQ4_XS on the smallest card that fits. Set a floor to hold the comparison at one level.
04

Rank by throughput rather than spec sheet

Sort by speed to see how cards rank for NVILA 15B. It will not match a gaming ordering — generation is bound by memory bandwidth, which is why the B200 tops it at 226 tok/s.
05

Check the fit verdict before buying

Tight means NVILA 15B 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

Each card page repeats this sweep for every model we hold. It answers what else the hardware is good for, beyond NVILA 15B.

Answers

NVILA 15B — common questions

Can I run NVILA 15B on a 16 GB GPU?

Yes. A Tesla V100 SXM2 16 GB with 16 GB runs it at Q6_K, using about 13.3 GB and generating roughly 46.4 tokens per second — a tight fit.

Can I run NVILA 15B on a 24 GB GPU?

Yes. A GeForce RTX 5090 D V2 with 24 GB runs it at Q8_0, using about 16.8 GB and generating roughly 37.8 tokens per second — a comfortable fit.

Is NVILA 15B open source?

Its weights are published, so NVILA 15B 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 NVILA 15B have?

NVILA 15B has 15B parameters. 15B. 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 NVILA 15B?

NVILA 15B was published by NVIDIA,Massachusetts Institute of Technology (MIT),University of California (UC) Berkeley,University of California San Diego,University of Washington,Tsinghua University, based in United States of America, categorised as industry,Academia,Academia,Academia,Academia,Academia.

When was NVILA 15B released?

NVILA 15B was published in December 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 NVILA 15B used for?

NVILA 15B works in Vision, Language, Multimodal, Video, and is recorded as handling visual question answering, Video description, Language modeling/generation, Question answering, Character recognition (OCR). These are the areas it was designed around; they describe intent rather than a hard boundary.

Where can I download NVILA 15B?

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

Can I run NVILA 15B 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 2.6 GB. Our figures for NVILA 15B assume it is fully resident.

Would two GPUs run NVILA 15B faster?

Two cards buy memory rather than speed. That matters for NVILA 15B only if one card cannot hold it — 306 can, so a second adds little.

Why does the quantisation differ between cards for NVILA 15B?

Because capacity varies, so does how hard NVILA 15B has to be squeezed — 4 distinct levels appear in the table above. Set a minimum quality to compare at one.

How accurate are these NVILA 15B speed estimates?

They are calculated from specifications rather than measured, and each carries a range — 136–361 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 NVILA 15B?

The smallest card in our catalogue that holds NVILA 15B is the P102-101, with 10 GB of memory. It runs the model at IQ4_XS using about 8.9 GB, and produces roughly 18.9 tokens per second. 306 cards in total can run it.

How fast is NVILA 15B on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 226 tokens per second; the slowest that still runs it manages considerably less. Reading speed is around ten tokens per second, and 266 of the cards that can run NVILA 15B clear that.

How much VRAM does NVILA 15B need?

About 8.9 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.

Can I run NVILA 15B on a 12 GB GPU?

Yes. A GeForce RTX 3080 Ti with 12 GB runs it at Q4_K_M, using about 9.8 GB and generating roughly 59.5 tokens per second — a tight fit.

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