code2seq TPS calculator

Open weights Technion - Israel Institute of Technology,Facebook AI Research 37M parameters February 2019

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 · 996 tok/s

Fastest card

B200

91,574 tok/s · 180 GB

Which GPUs can run code2seq?

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
91,574 tok/s 54,944–146,518 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	0.7 GB	Q8_0	Comfortable
91,574 tok/s 54,944–146,518 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	0.7 GB	Q8_0	Comfortable
73,124 tok/s 43,874–116,999 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	0.7 GB	Q8_0	Comfortable
73,124 tok/s 43,874–116,999 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	0.7 GB	Q8_0	Comfortable
58,481 tok/s 35,089–93,570 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	0.7 GB	Q8_0	Comfortable
55,975 tok/s 33,585–89,559 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	0.7 GB	Q8_0	Comfortable
55,975 tok/s 33,585–89,559 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	0.7 GB	Q8_0	Comfortable
53,571 tok/s 32,142–85,713 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	0.7 GB	Q8_0	Comfortable
47,544 tok/s 28,526–76,070 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	0.7 GB	Q8_0	Comfortable
47,544 tok/s 28,526–76,070 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	0.7 GB	Q8_0	Comfortable
47,544 tok/s 28,526–76,070 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	0.7 GB	Q8_0	Comfortable
45,100 tok/s 27,060–72,160 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
38,461 tok/s 23,077–61,538 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
38,461 tok/s 23,077–61,538 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	0.7 GB	Q8_0	Comfortable
38,461 tok/s 23,077–61,538 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
38,461 tok/s 23,077–61,538 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
38,461 tok/s 23,077–61,538 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
29,285 tok/s 17,571–46,857 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	0.7 GB	Q8_0	Comfortable
29,285 tok/s 17,571–46,857 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	0.7 GB	Q8_0	Comfortable
24,404 tok/s 14,643–39,047 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	0.7 GB	Q8_0	Comfortable
23,884 tok/s 14,330–38,214 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	0.7 GB	Q8_0	Comfortable
23,351 tok/s 14,011–37,362 · low confidence	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	0.7 GB	Q8_0	Comfortable
23,351 tok/s 14,011–37,362 · low confidence	A100X NVIDIA	80 GB	2,040 GB/s	Jun 2021	0.7 GB	Q8_0	Comfortable
23,351 tok/s 14,011–37,362 · low confidence	A800 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Aug 2022	0.7 GB	Q8_0	Comfortable
23,351 tok/s 14,011–37,362 · low confidence	H100 CNX NVIDIA	80 GB	2,040 GB/s	Mar 2023	0.7 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: Technion - Israel Institute of Technology,Facebook AI Research
Organisation type: Academia,Industry
Country: Israel, United States of America, France
Published: 21 February 2019
Authors: Uri Alon, Shaked Brody, Omer Levy, Eran Yahav

What it does

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

Domain: Language
Task: Language modeling

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: 37M
Training data: 46,033,536 tokens
Epochs: 52

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 × 10¹⁹ FLOP
How it was established: Operation counting

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.

Record confidence: Confident

Sources

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

Reference: code2seq: Generating Sequences from Structured Representations of Code
Last updated: 28 November 2025

The extremes

The ten fastest GPUs that run code2seq

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 code2seq

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

0.7 GB

Fastest

91,574 tok/s

code2seq is small enough at 37M 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 996 tokens per second.

A B200 is the fastest we calculate for it: about 91,574 tokens per second, from 8,000 GB/s of memory bandwidth.

Where it came from

code2seq was published by Technion - Israel Institute of Technology,Facebook AI Research, in Israel, in February 2019. academia,Industry is the category the publisher falls under.

It works in Language, and is recorded as doing language modeling.

The weights are published, so it can be downloaded and run on your own hardware indefinitely, offline, with no account attached.

Understanding the speeds

The median result is around 2,571.4 tokens per second; 818 cards produce text faster than most people read it.

Every weight participates in every token here, so bandwidth is the whole story: the ranking below is effectively a ranking of memory throughput.

Memory here is estimated from size rather than computed from the architecture, which is not recorded for this model — the numbers are indicative rather than exact.

What went into building it

Producing it required around 1.2 × 10¹⁹ FLOP of arithmetic, which is a statement about the training budget rather than about inference.

Around 46,033,536 tokens went into training it.

Step by step

How to choose a GPU for code2seq

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

Every card here has been checked against code2seq — around 0.7 GB at Q8_0. Capacity is the gate — a card either holds it or it does not.
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 code2seq 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 code2seq by squeezing it further than you would want.
04

Compare tokens per second, not specifications

Sort by speed to see how cards rank for code2seq. It will not match a gaming ordering — generation is bound by memory bandwidth, which is why the B200 tops it at 91,574 tok/s.
05

Read the fit column last

A tight fit runs code2seq 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

Every card name links to its own page, which runs the same calculation across the whole model catalogue. Worth a look before buying for code2seq alone — a card is usually bought for more than one model.

Answers

code2seq — common questions

How much VRAM does code2seq need?

About 0.7 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 code2seq on a 8 GB GPU?

Yes. A CMP 170HX 8 GB with 8 GB runs it at Q8_0, using about 0.7 GB and generating roughly 17,056 tokens per second — a comfortable fit.

Can I run code2seq on a 12 GB GPU?

Yes. A GeForce RTX 3080 Ti with 12 GB runs it at Q8_0, using about 0.7 GB and generating roughly 10,444 tokens per second — a comfortable fit.

Can I run code2seq on a 16 GB GPU?

Yes. A Tesla V100 SXM2 16 GB with 16 GB runs it at Q8_0, using about 0.7 GB and generating roughly 12,935 tokens per second — a comfortable fit.

Can I run code2seq on a 24 GB GPU?

Yes. A GeForce RTX 5090 D V2 with 24 GB runs it at Q8_0, using about 0.7 GB and generating roughly 15,339 tokens per second — a comfortable fit.

Is code2seq open source?

Its weights are published, so code2seq 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 code2seq have?

code2seq has 37M parameters. while our code2seq model had only 37M. 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 code2seq?

code2seq was published by Technion - Israel Institute of Technology,Facebook AI Research, based in Israel, categorised as academia,Industry.

When was code2seq released?

code2seq was published in February 2019. 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 code2seq used for?

code2seq works in Language, and is recorded as handling language modeling. Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.

Where can I download code2seq?

The weights for code2seq 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 code2seq?

Around 1.2 × 10¹⁹ FLOP. 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 code2seq if it does not fit in my GPU?

It can be split between the card and system memory, but code2seq generates painfully slowly that way. Nothing on this page assumes offloading.

Would two GPUs run code2seq faster?

A second card roughly doubles the memory available but not the generation rate. With 818 cards already able to run code2seq alone, the case for pairing is weak.

Why does the quantisation differ between cards for code2seq?

A larger card holds a more accurate copy. Across the cards that run code2seq, 1 compression levels are used; the floor control above pins it to one.

How accurate are these code2seq speed estimates?

They are calculated from specifications rather than measured, and each carries a range — 54,944–146,518 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 code2seq?

The smallest card in our catalogue that holds code2seq is the Tesla C1080, with 4 GB of memory. It runs the model at Q8_0 using about 0.7 GB, and produces roughly 996 tokens per second. 818 cards in total can run it.

How fast is code2seq on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 91,574 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 code2seq clear that.

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