CodeGen2 TPS calculator

Open weights Salesforce 16B parameters May 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

306 of 818 cards that can run it

Smallest card that fits

P102-101

10 GB · Q3_K_M · 19.5 tok/s

Fastest card

B200

212 tok/s · 180 GB

Which GPUs can run CodeGen2?

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
212 tok/s 127–339 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	17.8 GB	Q8_0	Comfortable
212 tok/s 127–339 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	17.8 GB	Q8_0	Comfortable
169 tok/s 101–271 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	17.8 GB	Q8_0	Comfortable
169 tok/s 101–271 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	17.8 GB	Q8_0	Comfortable
135 tok/s 81–216 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	17.8 GB	Q8_0	Comfortable
129 tok/s 78–207 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	17.8 GB	Q8_0	Comfortable
129 tok/s 78–207 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	17.8 GB	Q8_0	Comfortable
124 tok/s 74–198 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	17.8 GB	Q8_0	Comfortable
111 tok/s 67–178 · low confidence	CMP 170HX 10 GB NVIDIA	10 GB	1,560 GB/s	Sep 2021	8.5 GB	Q3_K_M	Tight
110 tok/s 66–176 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	17.8 GB	Q8_0	Comfortable
110 tok/s 66–176 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	17.8 GB	Q8_0	Comfortable
110 tok/s 66–176 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	17.8 GB	Q8_0	Comfortable
104 tok/s 63–167 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	17.8 GB	Q8_0	Comfortable
88.9 tok/s 53–142 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	17.8 GB	Q8_0	Comfortable
88.9 tok/s 53–142 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	17.8 GB	Q8_0	Comfortable
88.9 tok/s 53–142 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	17.8 GB	Q8_0	Comfortable
88.9 tok/s 53–142 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	17.8 GB	Q8_0	Comfortable
88.9 tok/s 53–142 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	17.8 GB	Q8_0	Comfortable
67.7 tok/s 41–108 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	17.8 GB	Q8_0	Comfortable
67.7 tok/s 41–108 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	17.8 GB	Q8_0	Comfortable
56.4 tok/s 34–90 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	17.8 GB	Q8_0	Comfortable
55.8 tok/s 33–89 · low confidence	GeForce RTX 3080 12 GB NVIDIA	12 GB	912 GB/s	Jan 2022	10.4 GB	Q4_K_M	Tight
55.8 tok/s 33–89 · low confidence	GeForce RTX 3080 Ti NVIDIA	12 GB	912 GB/s	May 2021	10.4 GB	Q4_K_M	Tight
55.2 tok/s 33–88 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	17.8 GB	Q8_0	Comfortable
54.3 tok/s 33–87 · low confidence	CMP 90HX NVIDIA	10 GB	760 GB/s	Jul 2021	8.5 GB	Q3_K_M	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: Salesforce
Organisation type: Industry
Country: United States of America
Published: 3 May 2023
Authors: Erik Nijkamp, Hiroaki Hayashi, Caiming Xiong, Silvio Savarese, Yingbo Zhou

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: 16B
Training data: tokens

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)

How it is classified

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

Likely above 10²³ FLOP: Yes
Record confidence: Confident
Citations: 235

Sources

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

Reference: CodeGen2: Lessons for Training LLMs on Programming and Natural Languages
Last updated: 25 May 2026

The extremes

The ten fastest GPUs for CodeGen2

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 CodeGen2

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

P102-101

Memory needed

8.5 GB

Fastest

212 tok/s

CodeGen2 is small enough at 16B parameters that hardware is rarely the obstacle — 306 of the cards we track can run it, including cards several years old.

The least hardware that works is a P102-101. Its 10 GB is enough at Q3_K_M compression, giving roughly 19.5 tokens per second.

The quickest result comes from a B200 at around 212 tokens per second — its 8,000 GB/s of bandwidth is what buys that.

Where it came from

CodeGen2 was published by Salesforce, in United States of America, in May 2023. The organisation is categorised as industry.

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

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 19.8 tokens per second; 259 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.

Step by step

How to choose a GPU for CodeGen2

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 CodeGen2 — around 8.5 GB at Q3_K_M. Capacity is the gate — a card either holds it or it does not.
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 CodeGen2 stops fitting a card that seemed fine.
03

Decide how much compression you will accept

The quantisation column varies by card, because a bigger card holds a more accurate copy of CodeGen2 — Q3_K_M on the smallest card that fits. Set a floor to hold the comparison at one level.
04

Sort by speed

Sort by speed to see how cards rank for CodeGen2. It will not match a gaming ordering — generation is bound by memory bandwidth, which is why the B200 tops it at 212 tok/s.
05

Check the fit verdict before buying

Tight means CodeGen2 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

Following a card through to its own page shows every other model it can hold, which is the question that follows once CodeGen2 is settled.

Answers

CodeGen2 — common questions

Can I run CodeGen2 on a 16 GB GPU?

Yes. A Tesla V100 SXM2 16 GB with 16 GB runs it at Q6_K, using about 14.1 GB and generating roughly 43.5 tokens per second — a tight fit.

Can I run CodeGen2 on a 24 GB GPU?

Yes. A GeForce RTX 5090 D V2 with 24 GB runs it at Q8_0, using about 17.8 GB and generating roughly 35.5 tokens per second — a comfortable fit.

Is CodeGen2 open source?

Its weights are published, so CodeGen2 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 CodeGen2 have?

CodeGen2 has 16B parameters. 16B for largest CodeGen2 model. 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 CodeGen2?

CodeGen2 was published by Salesforce, based in United States of America, categorised as industry.

When was CodeGen2 released?

CodeGen2 was published in May 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 CodeGen2 used for?

CodeGen2 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 CodeGen2?

The weights for CodeGen2 are published, though we do not hold a repository link for it. This site calculates hardware requirements rather than hosting model files.

Can I run CodeGen2 if it does not fit in my GPU?

It can be split between the card and system memory, but CodeGen2 generates painfully slowly that way — the nearest miss we calculate is short by 3.2 GB. Nothing on this page assumes offloading.

Would two GPUs run CodeGen2 faster?

Capacity adds across cards; throughput does not. Since 306 of the cards we track already hold CodeGen2 on their own, a second card is rarely the answer here.

Why does the quantisation differ between cards for CodeGen2?

Because capacity varies, so does how hard CodeGen2 has to be squeezed — 5 distinct levels appear in the table above. Set a minimum quality to compare at one.

How accurate are these CodeGen2 speed estimates?

These are estimates with real error bars. The fastest result here, 127–339 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 CodeGen2?

The smallest card in our catalogue that holds CodeGen2 is the P102-101, with 10 GB of memory. It runs the model at Q3_K_M using about 8.5 GB, and produces roughly 19.5 tokens per second. 306 cards in total can run it.

How fast is CodeGen2 on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 212 tokens per second; the slowest that still runs it manages considerably less. Reading speed is around ten tokens per second, and 259 of the cards that can run CodeGen2 clear that.

How much VRAM does CodeGen2 need?

About 8.5 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.

Can I run CodeGen2 on a 12 GB GPU?

Yes. A GeForce RTX 3080 Ti with 12 GB runs it at Q4_K_M, using about 10.4 GB and generating roughly 55.8 tokens per second — a tight 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