IDEFICS-80B TPS calculator

Open weights Hugging Face 80B parameters August 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

58 of 818 cards that can run it

Smallest card that fits

Quadro RTX 8000

48 GB · Q3_K_M · 9.6 tok/s

Fastest card

B200

42.4 tok/s · 180 GB

Which GPUs can run IDEFICS-80B?

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

58 cards match

Calculating

					Needs	Quantisation	Fit
42.4 tok/s 25–68 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	86.4 GB	Q8_0	Comfortable
42.4 tok/s 25–68 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	86.4 GB	Q8_0	Comfortable
33.8 tok/s 20–54 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	86.4 GB	Q8_0	Comfortable
33.8 tok/s 20–54 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	86.4 GB	Q8_0	Comfortable
30.3 tok/s 18–48 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	67.7 GB	Q6_K	Comfortable
27.1 tok/s 16–43 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	86.4 GB	Q8_0	Comfortable
26.7 tok/s 16–43 · low confidence	GRID A100B NVIDIA	48 GB	1,870 GB/s	May 2020	39.8 GB	Q3_K_M	Tight
25.9 tok/s 16–41 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	86.4 GB	Q8_0	Comfortable
25.9 tok/s 16–41 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	86.4 GB	Q8_0	Comfortable
25.9 tok/s 16–41 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	67.7 GB	Q6_K	Tight
25.9 tok/s 16–41 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	67.7 GB	Q6_K	Comfortable
25.9 tok/s 16–41 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	67.7 GB	Q6_K	Tight
24.8 tok/s 15–40 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	86.4 GB	Q8_0	Comfortable
24.7 tok/s 15–40 · low confidence	H100 SXM5 64 GB NVIDIA	64 GB	2,020 GB/s	Mar 2023	49.1 GB	Q4_K_M	Tight
22.0 tok/s 13–35 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	86.4 GB	Q8_0	Comfortable
22.0 tok/s 13–35 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	86.4 GB	Q8_0	Comfortable
22.0 tok/s 13–35 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	86.4 GB	Q8_0	Comfortable
19.1 tok/s 11–31 · low confidence	RTX PRO 5000 Blackwell NVIDIA	48 GB	1,340 GB/s	Mar 2025	39.8 GB	Q3_K_M	Tight
17.8 tok/s 11–28 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	86.4 GB	Q8_0	Tight
17.8 tok/s 11–28 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	86.4 GB	Q8_0	Tight
15.7 tok/s 9–25 · low confidence	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	67.7 GB	Q6_K	Tight
15.7 tok/s 9–25 · low confidence	A100X NVIDIA	80 GB	2,040 GB/s	Jun 2021	67.7 GB	Q6_K	Tight
15.7 tok/s 9–25 · low confidence	A800 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Aug 2022	67.7 GB	Q6_K	Tight
15.7 tok/s 9–25 · low confidence	H100 CNX NVIDIA	80 GB	2,040 GB/s	Mar 2023	67.7 GB	Q6_K	Tight
15.7 tok/s 9–25 · low confidence	H100 PCIe 80 GB NVIDIA	80 GB	2,040 GB/s	Oct 2022	67.7 GB	Q6_K	Tight

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: Hugging Face
Organisation type: Industry
Country: United States of America
Published: 22 August 2023
Authors: Hugo Laurencon, Daniel van Strien, Stas Bekman, Leo Tronchon, Lucile Saulnier, Thomas Wang, Siddharth Karamcheti, Amanpreet Singh, Giada Pistilli, Yacine Jernite, Victor Sanh

What it does

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

Domain: Multimodal, Language, Vision
Task: Language modeling, Image captioning, Visual question answering
Base model: LLaMA-65B,CLIP ViT-H/14 - LAION-2B
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: 80B
Training data: 149,600,000,000 tokens
Batch size: 3,670,000

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: 1.2 × 10²³ FLOP
How it was established: Hardware,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 A100
Chips used: 512
Wall-clock time: 672 hours (28 days)
Power draw: 407.0 kW
Cloud vendor: AWS

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: Unreleased
Hugging Face: HuggingFaceM4

How it is classified

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

Foundation model: Yes
Likely above 10²³ FLOP: Yes
Record confidence: Confident

Sources

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

Reference: Introducing IDEFICS: An Open Reproduction of State-of-the-Art Visual Language Model
Last updated: 28 November 2025

The extremes

The ten fastest GPUs for IDEFICS-80B

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 IDEFICS-80B

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

Quadro RTX 8000

Memory needed

39.8 GB

Fastest

42.4 tok/s

IDEFICS-80B sits at 80B parameters, which puts it above consumer hardware and into the range where a card is bought for this purpose rather than repurposed for it. 58 of the cards we track can hold it.

The entry point is the Quadro RTX 8000: 48 GB of memory, Q3_K_M compression, roughly 9.6 tokens per second.

The quickest result comes from a B200 at around 42.4 tokens per second — its 8,000 GB/s of bandwidth is what buys that.

What this model is

IDEFICS-80B was published by Hugging Face, in United States of America, in August 2023. industry is the category the publisher falls under.

It works in Multimodal, Language, Vision, and is recorded as doing language modeling, Image captioning, Visual question answering.

It builds on LLaMA-65B,CLIP ViT-H/14 - LAION-2B, which is why it shares that model's general shape and size.

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 HuggingFaceM4 organisation on Hugging Face.

What decides the speed

The median result is around 14.9 tokens per second; 43 cards produce text faster than most people read it.

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.

What went into building it

The training run consumed about 1.2 × 10²³ FLOP, on NVIDIA A100. That figure describes the cost of creating it and has no bearing on how quickly it generates text.

Around 149,600,000,000 tokens went into training it.

Step by step

How to choose a GPU for IDEFICS-80B

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

01

Read the memory figure first

The table lists every card that can hold IDEFICS-80B — around 39.8 GB at Q3_K_M. That figure, not the card's headline performance, is what decides whether it runs.
02

Set the context length you will work at

The conversation occupies memory too, and grows as it goes. Set the slider to the length you expect: at long context IDEFICS-80B can slip off a card that handles short questions easily.
03

Choose how far you will compress it

Compression is what makes IDEFICS-80B 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 IDEFICS-80B. It will not match a gaming ordering — generation is bound by memory bandwidth, which is why the B200 tops it at 42.4 tok/s.
05

Look at the headroom, not just the fit

Tight means IDEFICS-80B 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

Open the card you have settled on

Each card page repeats this sweep for every model we hold. It answers what else the hardware is good for, beyond IDEFICS-80B.

Answers

IDEFICS-80B — common questions

Is IDEFICS-80B open source?

Its weights are published, so IDEFICS-80B 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 IDEFICS-80B have?

IDEFICS-80B has 80B parameters. IDEFICS... comes in two variants—the base version and the instructed version. Each variant is available at the 9 billion and 80 billion parameter sizes. 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 IDEFICS-80B?

IDEFICS-80B was published by Hugging Face, based in United States of America, categorised as industry.

When was IDEFICS-80B released?

IDEFICS-80B was published in August 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 IDEFICS-80B used for?

IDEFICS-80B works in Multimodal, Language, Vision, and is recorded as handling language modeling, Image captioning, Visual question answering. 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 IDEFICS-80B?

Its weights are published under the HuggingFaceM4 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 IDEFICS-80B?

Around 1.2 × 10²³ FLOP, on NVIDIA A100. 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 IDEFICS-80B if it does not fit in my GPU?

It can be split between the card and system memory, but IDEFICS-80B generates painfully slowly that way — the nearest miss we calculate is short by 13.1 GB. Nothing on this page assumes offloading.

Would two GPUs run IDEFICS-80B faster?

Capacity adds across cards; throughput does not. Since 58 of the cards we track already hold IDEFICS-80B on their own, a second card is rarely the answer here.

Why does the quantisation differ between cards for IDEFICS-80B?

Each card is shown running the least-compressed copy it can hold, and IDEFICS-80B appears at 5 different compression levels across the cards that fit it. Bigger cards get the more accurate version.

How accurate are these IDEFICS-80B speed estimates?

These are estimates with real error bars. The fastest result here, 25–68 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 IDEFICS-80B?

The smallest card in our catalogue that holds IDEFICS-80B is the Quadro RTX 8000, with 48 GB of memory. It runs the model at Q3_K_M using about 39.8 GB, and produces roughly 9.6 tokens per second. 58 cards in total can run it.

How fast is IDEFICS-80B on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 42.4 tokens per second; the slowest that still runs it manages considerably less. Reading speed is around ten tokens per second, and 43 of the cards that can run IDEFICS-80B clear that.

How much VRAM does IDEFICS-80B need?

About 39.8 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 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