CodeT5-base TPS calculator

Open weights Salesforce,Nanyang Technological University 220M parameters November 2021

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 cards that can run it

818 cards we hold specifications for

Smallest card that fits

Tesla C1080

4 GB · Q8_0 · 168 tok/s

Fastest card

B200

15,401 tok/s · 180 GB

Which GPUs can run CodeT5-base?

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
15,401 tok/s 9,241–24,642 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	0.9 GB	Q8_0	Comfortable
15,401 tok/s 9,241–24,642 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	0.9 GB	Q8_0	Comfortable
12,298 tok/s 7,379–19,677 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	0.9 GB	Q8_0	Comfortable
12,298 tok/s 7,379–19,677 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	0.9 GB	Q8_0	Comfortable
9,836 tok/s 5,901–15,737 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	0.9 GB	Q8_0	Comfortable
9,414 tok/s 5,648–15,062 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	0.9 GB	Q8_0	Comfortable
9,414 tok/s 5,648–15,062 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	0.9 GB	Q8_0	Comfortable
9,010 tok/s 5,406–14,415 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	0.9 GB	Q8_0	Comfortable
7,996 tok/s 4,798–12,794 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	0.9 GB	Q8_0	Comfortable
7,996 tok/s 4,798–12,794 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	0.9 GB	Q8_0	Comfortable
7,996 tok/s 4,798–12,794 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	0.9 GB	Q8_0	Comfortable
7,585 tok/s 4,551–12,136 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	0.9 GB	Q8_0	Comfortable
6,468 tok/s 3,881–10,350 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	0.9 GB	Q8_0	Comfortable
6,468 tok/s 3,881–10,350 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	0.9 GB	Q8_0	Comfortable
6,468 tok/s 3,881–10,350 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	0.9 GB	Q8_0	Comfortable
6,468 tok/s 3,881–10,350 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	0.9 GB	Q8_0	Comfortable
6,468 tok/s 3,881–10,350 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	0.9 GB	Q8_0	Comfortable
4,925 tok/s 2,955–7,880 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	0.9 GB	Q8_0	Comfortable
4,925 tok/s 2,955–7,880 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	0.9 GB	Q8_0	Comfortable
4,104 tok/s 2,463–6,567 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	0.9 GB	Q8_0	Comfortable
4,017 tok/s 2,410–6,427 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	0.9 GB	Q8_0	Comfortable
3,927 tok/s 2,356–6,284 · low confidence	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	0.9 GB	Q8_0	Comfortable
3,927 tok/s 2,356–6,284 · low confidence	A100X NVIDIA	80 GB	2,040 GB/s	Jun 2021	0.9 GB	Q8_0	Comfortable
3,927 tok/s 2,356–6,284 · low confidence	A800 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Aug 2022	0.9 GB	Q8_0	Comfortable
3,927 tok/s 2,356–6,284 · low confidence	H100 CNX NVIDIA	80 GB	2,040 GB/s	Mar 2023	0.9 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: Salesforce,Nanyang Technological University
Organisation type: Industry,Academia
Country: United States of America, Singapore
Published: 1 November 2021
Authors: Yue Wang, Weishi Wang, Shafiq Joty, Steven C.H. Hoi

What it does

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

Domain: Language
Task: Code generation
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: 220M
Training data: tokens
Epochs: 150

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: 1.6 × 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 A100
Wall-clock time: 288 hours (12 days)
Compute cost: $3,115

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

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: Likely
Citations: 1,964

Sources

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

Reference: CodeT5: Identifier-aware Unified Pre-trained Encoder-Decoder Models for Code Understanding and Generation
Last updated: 1 January 2026

The extremes

The ten fastest GPUs that run CodeT5-base

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 CodeT5-base

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

Tesla C1080

Memory needed

0.9 GB

Fastest

15,401 tok/s

CodeT5-base is small enough at 220M parameters that hardware is rarely the obstacle — 818 of the cards we track can run it, including cards several years old.

The entry point is the Tesla C1080: 4 GB of memory, Q8_0 compression, roughly 168 tokens per second.

At the other end, a B200 generates roughly 15,401 tokens per second on it, on the strength of 8,000 GB/s of memory bandwidth.

What this model is

CodeT5-base was published by Salesforce,Nanyang Technological University, in United States of America, in November 2021. industry,Academia is the category the publisher falls under.

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

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.

What decides the speed

The median result is around 432.5 tokens per second; 818 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.

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 1.6 × 10²¹ FLOP, on NVIDIA A100. 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 CodeT5-base

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 CodeT5-base actually needs — around 0.9 GB at Q8_0. No amount of processing power compensates for a card that cannot hold it.
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 CodeT5-base stops fitting a card that seemed fine.
03

Decide how much compression you will accept

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 CodeT5-base by squeezing it further than you would want.
04

Rank by throughput rather than spec sheet

The speed ordering for CodeT5-base is effectively an ordering by memory bandwidth, which is why the B200 tops it at 15,401 tok/s.
05

Look at the headroom, not just the fit

A tight fit runs CodeT5-base but leaves nothing spare for a longer conversation; comfortable has headroom. If you expect to grow the context, buy for comfortable.
06

See what else that card runs

Following a card through to its own page shows every other model it can hold, which is the question that follows once CodeT5-base is settled.

Answers

CodeT5-base — common questions

Can I run CodeT5-base on a 8 GB GPU?

Yes. A CMP 170HX 8 GB with 8 GB runs it at Q8_0, using about 0.9 GB and generating roughly 2,868 tokens per second — a comfortable fit.

Can I run CodeT5-base on a 12 GB GPU?

Yes. A GeForce RTX 3080 Ti with 12 GB runs it at Q8_0, using about 0.9 GB and generating roughly 1,756 tokens per second — a comfortable fit.

Can I run CodeT5-base on a 16 GB GPU?

Yes. A Tesla V100 SXM2 16 GB with 16 GB runs it at Q8_0, using about 0.9 GB and generating roughly 2,175 tokens per second — a comfortable fit.

Can I run CodeT5-base on a 24 GB GPU?

Yes. A GeForce RTX 5090 D V2 with 24 GB runs it at Q8_0, using about 0.9 GB and generating roughly 2,580 tokens per second — a comfortable fit.

Is CodeT5-base open source?

Its weights are published, so CodeT5-base 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 CodeT5-base have?

CodeT5-base has 220M parameters. "We build CodeT5 based on Huggingface’s T5 (Raffel et al., 2020) PyTorch implementation and employ two sizes of CodeT5-small (60M) and CodeT5-base (220M)". 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 CodeT5-base?

CodeT5-base was published by Salesforce,Nanyang Technological University, based in United States of America, categorised as industry,Academia.

When was CodeT5-base released?

CodeT5-base was published in November 2021. 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 CodeT5-base used for?

CodeT5-base works in Language, and is recorded as handling code generation. 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 CodeT5-base?

The weights for CodeT5-base 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 CodeT5-base?

Around 1.6 × 10²¹ 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 CodeT5-base if it does not fit in my GPU?

It can be split between the card and system memory, but CodeT5-base generates painfully slowly that way. Nothing on this page assumes offloading.

Would two GPUs run CodeT5-base faster?

A second card roughly doubles the memory available but not the generation rate. With 818 cards already able to run CodeT5-base alone, the case for pairing is weak.

Why does the quantisation differ between cards for CodeT5-base?

Because capacity varies, so does how hard CodeT5-base has to be squeezed — 1 distinct levels appear in the table above. Set a minimum quality to compare at one.

How accurate are these CodeT5-base speed estimates?

They are calculated from specifications rather than measured, and each carries a range — 9,241–24,642 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 CodeT5-base?

The smallest card in our catalogue that holds CodeT5-base is the Tesla C1080, with 4 GB of memory. It runs the model at Q8_0 using about 0.9 GB, and produces roughly 168 tokens per second. 818 cards in total can run it.

How fast is CodeT5-base on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 15,401 tokens per second; the slowest that still runs it manages considerably less. Reading speed is around ten tokens per second, and 818 of the cards that can run CodeT5-base clear that.

How much VRAM does CodeT5-base need?

About 0.9 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.

Source

Original publication

Record last updated 1 January 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