CodeGen-Mono 16.1B TPS calculator

Open weights Salesforce 16.1B parameters February 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 cards that can run it

818 cards we hold specifications for

Smallest card that fits

P102-101

10 GB · Q3_K_M · 19.3 tok/s

Fastest card

B200

210 tok/s · 180 GB

Which GPUs can run CodeGen-Mono 16.1B?

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
210 tok/s 126–337 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	17.9 GB	Q8_0	Comfortable
210 tok/s 126–337 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	17.9 GB	Q8_0	Comfortable
168 tok/s 101–269 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	17.9 GB	Q8_0	Comfortable
168 tok/s 101–269 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	17.9 GB	Q8_0	Comfortable
134 tok/s 81–215 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	17.9 GB	Q8_0	Comfortable
129 tok/s 77–206 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	17.9 GB	Q8_0	Comfortable
129 tok/s 77–206 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	17.9 GB	Q8_0	Comfortable
123 tok/s 74–197 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	17.9 GB	Q8_0	Comfortable
111 tok/s 66–177 · low confidence	CMP 170HX 10 GB NVIDIA	10 GB	1,560 GB/s	Sep 2021	8.6 GB	Q3_K_M	Tight
109 tok/s 66–175 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	17.9 GB	Q8_0	Comfortable
109 tok/s 66–175 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	17.9 GB	Q8_0	Comfortable
109 tok/s 66–175 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	17.9 GB	Q8_0	Comfortable
104 tok/s 62–166 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	17.9 GB	Q8_0	Comfortable
88.4 tok/s 53–141 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	17.9 GB	Q8_0	Comfortable
88.4 tok/s 53–141 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	17.9 GB	Q8_0	Comfortable
88.4 tok/s 53–141 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	17.9 GB	Q8_0	Comfortable
88.4 tok/s 53–141 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	17.9 GB	Q8_0	Comfortable
88.4 tok/s 53–141 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	17.9 GB	Q8_0	Comfortable
67.3 tok/s 40–108 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	17.9 GB	Q8_0	Comfortable
67.3 tok/s 40–108 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	17.9 GB	Q8_0	Comfortable
56.1 tok/s 34–90 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	17.9 GB	Q8_0	Comfortable
55.4 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
55.4 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
54.9 tok/s 33–88 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	17.9 GB	Q8_0	Comfortable
54.0 tok/s 32–86 · low confidence	CMP 90HX NVIDIA	10 GB	760 GB/s	Jul 2021	8.6 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: 27 February 2023
Authors: Erik Nijkamp, Bo Pang, Hiroaki Hayashi, Lifu Tu, Huan Wang, Yingbo Zhou, Silvio Savarese, Caiming Xiong

What it does

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

Domain: Language
Task: Code generation, Code autocompletion

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: 16.1B
Training data: 71,700,000,000 tokens
Epochs: 4.18

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: Google TPU v4

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: Likely
Citations: 1,473

Sources

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

Reference: CodeGen: An Open Large Language Model for Code with Multi-Turn Program Synthesis
Last updated: 25 May 2026

The extremes

The ten fastest GPUs that run CodeGen-Mono 16.1B

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 CodeGen-Mono 16.1B

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 it takes to run this model

Minimum card

P102-101

Memory needed

8.6 GB

Fastest

210 tok/s

CodeGen-Mono 16.1B reaches a parameter count of 16.1B. That is small enough that hardware is rarely the obstacle, including on cards several years old. The number of cards we track that can run it: 306.

The smallest card that holds it is P102-101, with a memory capacity of 10 GB, running it at a compression of Q3_K_M and producing around 19.3 tokens per second.

The quickest result comes from B200, generating roughly 210 tokens per second on the strength of a memory bandwidth of 8,000 GB/s.

Background

CodeGen-Mono 16.1B was published by Salesforce, in the country recorded as United States of America, during February 2023. The publishing organisation is categorised as industry.

It works in the domain of Language, and is recorded as performing the task of code generation, Code autocompletion.

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 19.6 tokens per second. Exceeding reading speed outright: 259 of them.

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.

Training and provenance

The training set ran to roughly 71,700,000,000 tokens of text.

Step by step

How to choose a GPU for CodeGen-Mono 16.1B

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

01

Start from the memory column

Every card here has been checked against CodeGen-Mono 16.1B, needing around 8.6 GB at a compression of Q3_K_M. Capacity is the gate — a card either holds it or it does not.
02

Match the context to your actual use

The conversation occupies memory too, and grows as it goes. Set the slider to the length you expect, because at long context a card that handles short questions easily can be dropped by CodeGen-Mono 16.1B.
03

Choose how far you will compress it

Compression is what makes a model fit smaller cards, at some cost in accuracy, reaching a compression of Q3_K_M on the smallest card that fits. Setting a minimum quality drops the cards that only manage it by squeezing further than you would want, and holds the comparison at one level.
04

Compare tokens per second, not specifications

The speed ordering is effectively an ordering by memory bandwidth, for CodeGen-Mono 16.1B. It will not match a gaming ordering, because generation is bound by memory bandwidth. The card topping the list is B200, at 210 tok/s.
05

Look at the headroom, not just the fit

Tight means it loads and works with no room to raise the context later, in the case of CodeGen-Mono 16.1B. Comfortable means you can grow the context later. That difference matters more than a few tokens per second, so buy for comfortable if you expect to.
06

Open the card you have settled on

Every card name links to its own page, which runs the same calculation across the whole model catalogue. A card is usually bought for more than one model, so it is worth a look before buying for CodeGen-Mono 16.1B.

Answers

CodeGen-Mono 16.1B — common questions

CodeGen-Mono 16.1B— can I run it if it does not fit in my GPU?

Only by offloading, which is usually a false economy: the part held in system memory drags the whole thing down. The nearest miss we calculate falls short by 3.2 GB. Every figure here assumes the whole model is resident on the card.

CodeGen-Mono 16.1B— would two GPUs run it faster?

Two cards buy memory rather than speed, which matters only if one card cannot hold it. The number that can: 306. So a second card is rarely the answer here.

CodeGen-Mono 16.1B— why does the quantisation differ between cards?

Because capacity varies, so does how hard it has to be squeezed. The number of distinct levels in the table above: 5. Bigger cards get the more accurate version, and the quality floor above pins the comparison to one level.

CodeGen-Mono 16.1B— how accurate are these speed estimates?

They are calculated from specifications rather than measured, and each carries a range. One example: 126–337 tok/s on B200. The same model and card vary by thirty to fifty per cent depending on the inference software and its version.

CodeGen-Mono 16.1B— what GPU do I need to run it?

The smallest card in our catalogue that holds it is P102-101, with a memory capacity of 10 GB. It runs the model at a compression of Q3_K_M using about 8.6 GB, and produces roughly 19.3 tokens per second. The number of cards able to run it in total: 306.

CodeGen-Mono 16.1B— how fast is it on a GPU?

It depends on the card. The quickest we calculate is B200, at about 210 tokens per second; the slowest that still runs it manages considerably less. Reading speed is around ten tokens per second, and the number of cards clearing that: 259.

CodeGen-Mono 16.1B— how much VRAM does it need?

It needs about 8.6 GB at a compression of Q3_K_M, 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.

CodeGen-Mono 16.1B— can I run it on a GPU holding 12 GB?

Yes. The card GeForce RTX 3080 Ti, holding 12 GB, runs it at a compression of Q4_K_M, using about 10.4 GB and generating roughly 55.4 tokens per second. The fit is tight.

CodeGen-Mono 16.1B— can I run it on a GPU holding 16 GB?

Yes. The card Tesla V100 SXM2 16 GB, holding 16 GB, runs it at a compression of Q6_K, using about 14.2 GB and generating roughly 43.2 tokens per second. The fit is tight.

CodeGen-Mono 16.1B— can I run it on a GPU holding 24 GB?

Yes. The card GeForce RTX 5090 D V2, holding 24 GB, runs it at a compression of Q8_0, using about 17.9 GB and generating roughly 35.3 tokens per second. The fit is comfortable.

CodeGen-Mono 16.1B— is it open source?

Its weights are published, so it 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.

CodeGen-Mono 16.1B— how many parameters does it have?

It has a parameter count of 16.1B. 16.1B parameters. 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.

CodeGen-Mono 16.1B— who created it?

It was published by Salesforce, based in United States of America, an organisation categorised as industry.

CodeGen-Mono 16.1B— when was it released?

It was published in February 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.

CodeGen-Mono 16.1B— what is it used for?

It works in the domain of Language, and is recorded as handling the task of code generation, Code autocompletion. 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.

CodeGen-Mono 16.1B— where can I download it?

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

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