SantaCoder TPS calculator

Open weights Hugging Face,ServiceNow,Massachusetts Institute of Technology (MIT),Wellesley College,Saama,EleutherAI,Huawei Noah's Ark Lab,Carnegie Mellon University (CMU) 1.1B parameters January 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

818 of 818 cards that can run it

Smallest card that fits

Tesla C1080

4 GB · Q8_0 · 33.5 tok/s

Fastest card

B200

3,080 tok/s · 180 GB

Which GPUs can run SantaCoder?

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
3,080 tok/s 1,848–4,928 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	1.9 GB	Q8_0	Comfortable
3,080 tok/s 1,848–4,928 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	1.9 GB	Q8_0	Comfortable
2,460 tok/s 1,476–3,935 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	1.9 GB	Q8_0	Comfortable
2,460 tok/s 1,476–3,935 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	1.9 GB	Q8_0	Comfortable
1,967 tok/s 1,180–3,147 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	1.9 GB	Q8_0	Comfortable
1,883 tok/s 1,130–3,012 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	1.9 GB	Q8_0	Comfortable
1,883 tok/s 1,130–3,012 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	1.9 GB	Q8_0	Comfortable
1,802 tok/s 1,081–2,883 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	1.9 GB	Q8_0	Comfortable
1,599 tok/s 960–2,559 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	1.9 GB	Q8_0	Comfortable
1,599 tok/s 960–2,559 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	1.9 GB	Q8_0	Comfortable
1,599 tok/s 960–2,559 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	1.9 GB	Q8_0	Comfortable
1,517 tok/s 910–2,427 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	1.9 GB	Q8_0	Comfortable
1,294 tok/s 776–2,070 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	1.9 GB	Q8_0	Comfortable
1,294 tok/s 776–2,070 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	1.9 GB	Q8_0	Comfortable
1,294 tok/s 776–2,070 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	1.9 GB	Q8_0	Comfortable
1,294 tok/s 776–2,070 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	1.9 GB	Q8_0	Comfortable
1,294 tok/s 776–2,070 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	1.9 GB	Q8_0	Comfortable
985 tok/s 591–1,576 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	1.9 GB	Q8_0	Comfortable
985 tok/s 591–1,576 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	1.9 GB	Q8_0	Comfortable
821 tok/s 493–1,313 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	1.9 GB	Q8_0	Comfortable
803 tok/s 482–1,285 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	1.9 GB	Q8_0	Comfortable
785 tok/s 471–1,257 · low confidence	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	1.9 GB	Q8_0	Comfortable
785 tok/s 471–1,257 · low confidence	A100X NVIDIA	80 GB	2,040 GB/s	Jun 2021	1.9 GB	Q8_0	Comfortable
785 tok/s 471–1,257 · low confidence	A800 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Aug 2022	1.9 GB	Q8_0	Comfortable
785 tok/s 471–1,257 · low confidence	H100 CNX NVIDIA	80 GB	2,040 GB/s	Mar 2023	1.9 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: Hugging Face,ServiceNow,Massachusetts Institute of Technology (MIT),Wellesley College,Saama,EleutherAI,Huawei Noah's Ark Lab,Carnegie Mellon University (CMU)
Organisation type: Industry,Industry,Academia,Academia,Research collective,Industry,Academia
Country: United States of America, China
Published: 9 January 2023
Authors: Loubna Ben Allal, Raymond Li, Denis Kocetkov, Chenghao Mou, Christopher Akiki, Carlos Munoz Ferrandis, Niklas Muennighoff, Mayank Mishra, Alex Gu, Manan Dey, Logesh Kumar Umapathi, Carolyn Jane Anderson, Yangtian Zi, Joel Lamy Poirier, Hailey Schoelkopf, Sergey Troshin, Dmitry Abulkhanov, Manuel Romero, Michael Lappert, Francesco De Toni, Bernardo García del Río, Qian Liu, Shamik Bose, Urvashi Bha…

What it does

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

Domain: Language
Task: Code generation

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: 1.1B
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: 2.1 × 10²¹ FLOP
How it was established: Hardware

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 V100
Wall-clock time: 150 hours

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 (restricted use)
Training code: Unreleased

How it is classified

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

Foundation model: Yes
Record confidence: Confident
Citations: 238

Sources

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

Reference: SantaCoder: don't reach for the stars!
Last updated: 25 May 2026

The extremes

The ten fastest GPUs for SantaCoder

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 SantaCoder

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

What you need to run it

Minimum card

Tesla C1080

Memory needed

1.9 GB

Fastest

3,080 tok/s

SantaCoder is small enough at 1.1B parameters that hardware is rarely the obstacle — 818 of the cards we track can run it, including cards several years old.

The least hardware that works is a Tesla C1080. Its 4 GB is enough at Q8_0 compression, giving roughly 33.5 tokens per second.

The quickest result comes from a B200 at around 3,080 tokens per second — its 8,000 GB/s of bandwidth is what buys that.

Background

SantaCoder was published by Hugging Face,ServiceNow,Massachusetts Institute of Technology (MIT),Wellesley College,Saama,EleutherAI,Huawei Noah's Ark Lab,Carnegie Mellon University (CMU), in United States of America, in January 2023. It comes out of industry,Industry,Academia,Academia,Research collective,Industry,Academia.

It works in Language, and is recorded as doing code generation.

Because its weights were released, nothing about running it depends on a provider staying available — it is yours once downloaded.

Reading the throughput figures

Half the cards that hold it manage more than 86.5 tokens per second, and 799 exceed reading speed outright.

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.

What went into building it

Training it took roughly 2.1 × 10²¹ FLOP of computation, on NVIDIA V100 — a measure of what producing the model cost, not of how fast it answers.

Step by step

How to choose a GPU for SantaCoder

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

Look at what SantaCoder actually needs — around 1.9 GB at Q8_0. No amount of processing power compensates for a card that cannot hold it.
02

Set the context length you will work at

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

Set a quality floor

Each card runs the least-compressed copy it can hold — Q8_0 on the smallest card that fits. Setting a floor drops the cards that only manage SantaCoder by squeezing it further than you would want.
04

Sort by speed

Ranking by tokens per second for SantaCoder follows memory bandwidth, not core counts, which is why the B200 tops it at 3,080 tok/s.
05

Check the fit verdict before buying

Tight means SantaCoder 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

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

Answers

SantaCoder — common questions

Can I run SantaCoder on a 12 GB GPU?

Yes. A GeForce RTX 3080 Ti with 12 GB runs it at Q8_0, using about 1.9 GB and generating roughly 351 tokens per second — a comfortable fit.

Can I run SantaCoder on a 16 GB GPU?

Yes. A Tesla V100 SXM2 16 GB with 16 GB runs it at Q8_0, using about 1.9 GB and generating roughly 435 tokens per second — a comfortable fit.

Can I run SantaCoder on a 24 GB GPU?

Yes. A GeForce RTX 5090 D V2 with 24 GB runs it at Q8_0, using about 1.9 GB and generating roughly 516 tokens per second — a comfortable fit.

Is SantaCoder open source?

Its weights are published, so SantaCoder 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 SantaCoder have?

SantaCoder has 1.1B parameters. 1.1B. 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 SantaCoder?

SantaCoder was published by Hugging Face,ServiceNow,Massachusetts Institute of Technology (MIT),Wellesley College,Saama,EleutherAI,Huawei Noah's Ark Lab,Carnegie Mellon University (CMU), based in United States of America, categorised as industry,Industry,Academia,Academia,Research collective,Industry,Academia.

When was SantaCoder released?

SantaCoder was published in January 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 SantaCoder used for?

SantaCoder works in Language, and is recorded as handling code generation. Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.

Where can I download SantaCoder?

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

How much compute was used to train SantaCoder?

Around 2.1 × 10²¹ FLOP, on NVIDIA V100. 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 SantaCoder 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 SantaCoder assume it is fully resident.

Would two GPUs run SantaCoder faster?

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

Why does the quantisation differ between cards for SantaCoder?

Each card is shown running the least-compressed copy it can hold, and SantaCoder appears at 1 different compression levels across the cards that fit it. Bigger cards get the more accurate version.

How accurate are these SantaCoder speed estimates?

Every figure is derived from memory bandwidth and model size, not benchmarked. That is why each is published as a range such as 1,848–4,928 tok/s on the B200 rather than a single number.

What GPU do I need to run SantaCoder?

The smallest card in our catalogue that holds SantaCoder is the Tesla C1080, with 4 GB of memory. It runs the model at Q8_0 using about 1.9 GB, and produces roughly 33.5 tokens per second. 818 cards in total can run it.

How fast is SantaCoder on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 3,080 tokens per second; the slowest that still runs it manages considerably less. Reading speed is around ten tokens per second, and 799 of the cards that can run SantaCoder clear that.

How much VRAM does SantaCoder need?

About 1.9 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 SantaCoder on a 8 GB GPU?

Yes. A CMP 170HX 8 GB with 8 GB runs it at Q8_0, using about 1.9 GB and generating roughly 574 tokens per second — a comfortable fit.

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