Kevin-32B TPS calculator

Open weights Cognition,Stanford University 32B parameters May 2025

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

132 cards that can run it

818 cards we hold specifications for

Smallest card that fits

RTX A4500

20 GB · Q3_K_M · 22.9 tok/s

Fastest card

B200

106 tok/s · 180 GB

Which GPUs can run Kevin-32B?

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

132 cards match

Calculating

					Needs	Quantisation	Fit
106 tok/s 64–169 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	35.0 GB	Q8_0	Comfortable
106 tok/s 64–169 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	35.0 GB	Q8_0	Comfortable
84.6 tok/s 51–135 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	35.0 GB	Q8_0	Comfortable
84.6 tok/s 51–135 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	35.0 GB	Q8_0	Comfortable
67.6 tok/s 41–108 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	35.0 GB	Q8_0	Comfortable
64.7 tok/s 39–104 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	35.0 GB	Q8_0	Comfortable
64.7 tok/s 39–104 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	35.0 GB	Q8_0	Comfortable
61.9 tok/s 37–99 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	35.0 GB	Q8_0	Comfortable
55.0 tok/s 33–88 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	35.0 GB	Q8_0	Comfortable
55.0 tok/s 33–88 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	35.0 GB	Q8_0	Comfortable
55.0 tok/s 33–88 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	35.0 GB	Q8_0	Comfortable
52.2 tok/s 31–83 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	35.0 GB	Q8_0	Comfortable
44.5 tok/s 27–71 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	35.0 GB	Q8_0	Comfortable
44.5 tok/s 27–71 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	35.0 GB	Q8_0	Comfortable
44.5 tok/s 27–71 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	35.0 GB	Q8_0	Comfortable
44.5 tok/s 27–71 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	35.0 GB	Q8_0	Comfortable
44.5 tok/s 27–71 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	35.0 GB	Q8_0	Comfortable
40.9 tok/s 25–66 · low confidence	GeForce RTX 5090 D V2 NVIDIA	24 GB	1,340 GB/s	Aug 2025	20.1 GB	Q4_K_M	Tight
37.3 tok/s 22–60 · low confidence	A30X NVIDIA	24 GB	1,220 GB/s	Apr 2021	20.1 GB	Q4_K_M	Tight
36.0 tok/s 22–58 · low confidence	DRIVE A100 PROD NVIDIA	32 GB	1,870 GB/s	May 2020	27.5 GB	Q6_K	Tight
36.0 tok/s 22–58 · low confidence	GRID A100A NVIDIA	32 GB	1,870 GB/s	May 2020	27.5 GB	Q6_K	Tight
34.4 tok/s 21–55 · low confidence	GeForce RTX 5090 NVIDIA	32 GB	1,790 GB/s	Jan 2025	27.5 GB	Q6_K	Tight
34.4 tok/s 21–55 · low confidence	GeForce RTX 5090 D NVIDIA	32 GB	1,790 GB/s	Jan 2025	27.5 GB	Q6_K	Tight
33.9 tok/s 20–54 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	35.0 GB	Q8_0	Comfortable
33.9 tok/s 20–54 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	35.0 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: Cognition,Stanford University
Organisation type: Industry,Academia
Country: United States of America
Published: 6 May 2025
Authors: Carlo Baronio, Pietro Marsella, Ben Pan, Silas Alberti

What it does

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

Domain: Language
Task: Code generation
Base model: QwQ-32B

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: 32B
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 (non-commercial)
Training code: Unreleased
Hugging Face: cognition-ai

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: Kevin-32B: Multi-Turn RL for Writing CUDA Kernels
Last updated: 28 November 2025

The extremes

The ten fastest GPUs that run Kevin-32B

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 Kevin-32B

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

RTX A4500

Memory needed

16.3 GB

Fastest

106 tok/s

With 32B parameters, Kevin-32B lands in the range a serious desktop card can handle once the weights are compressed. 132 of the cards we track can run it.

The entry point is the RTX A4500: 20 GB of memory, Q3_K_M compression, roughly 22.9 tokens per second.

A B200 is the fastest we calculate for it: about 106 tokens per second, from 8,000 GB/s of memory bandwidth.

What this model is

Kevin-32B was published by Cognition,Stanford University, in United States of America, in May 2025. industry,Academia is the category the publisher falls under.

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

It builds on QwQ-32B, which is why it shares that model's general shape and size.

Published weights mean the model runs on your machine rather than someone else's, which is what makes the hardware question below answerable at all. It is published under the cognition-ai organisation on Hugging Face.

What decides the speed

Across every card that can run it, the middle of the range is about 20.7 tokens per second, and 103 of them clear the ten tokens per second that roughly matches reading speed.

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 Kevin-32B

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 Kevin-32B actually needs — around 16.3 GB at Q3_K_M. No amount of processing power compensates for a card that cannot hold it.
02

Set the context length you will work at

The conversation occupies memory too, and grows as it goes. Set the slider to the length you expect: at long context Kevin-32B can slip off a card that handles short questions easily.
03

Set a quality floor

The quantisation column varies by card, because a bigger card holds a more accurate copy of Kevin-32B — Q3_K_M on the smallest card that fits. Set a floor to hold the comparison at one level.
04

Rank by throughput rather than spec sheet

Ranking by tokens per second for Kevin-32B follows memory bandwidth, not core counts, which is why the B200 tops it at 106 tok/s.
05

Look at the headroom, not just the fit

The fit column separates cards that just manage Kevin-32B from those with room to spare. Buy for the second if the context might grow.
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 Kevin-32B is settled.

Answers

Kevin-32B — common questions

Can I run Kevin-32B if it does not fit in my GPU?

Partly. Layers that do not fit sit in system memory and run at a fraction of the speed, so a mostly-offloaded Kevin-32B is rarely worth using — the nearest miss we calculate is short by 5.7 GB. Every figure here assumes the whole model is on the card.

Would two GPUs run Kevin-32B faster?

Capacity adds across cards; throughput does not. Since 132 of the cards we track already hold Kevin-32B on their own, a second card is rarely the answer here.

Why does the quantisation differ between cards for Kevin-32B?

A larger card holds a more accurate copy. Across the cards that run Kevin-32B, 5 compression levels are used; the floor control above pins it to one.

How accurate are these Kevin-32B 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 64–169 tok/s on the B200 rather than a single number.

What GPU do I need to run Kevin-32B?

The smallest card in our catalogue that holds Kevin-32B is the RTX A4500, with 20 GB of memory. It runs the model at Q3_K_M using about 16.3 GB, and produces roughly 22.9 tokens per second. 132 cards in total can run it.

How fast is Kevin-32B on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 106 tokens per second; the slowest that still runs it manages considerably less. Reading speed is around ten tokens per second, and 103 of the cards that can run Kevin-32B clear that.

How much VRAM does Kevin-32B need?

About 16.3 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 Kevin-32B on a 24 GB GPU?

Yes. A GeForce RTX 5090 D V2 with 24 GB runs it at Q4_K_M, using about 20.1 GB and generating roughly 40.9 tokens per second — a tight fit.

Is Kevin-32B open source?

Its weights are published, so Kevin-32B 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 Kevin-32B have?

Kevin-32B has 32B parameters. 32B. 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 Kevin-32B?

Kevin-32B was published by Cognition,Stanford University, based in United States of America, categorised as industry,Academia.

When was Kevin-32B released?

Kevin-32B was published in May 2025.

What is Kevin-32B used for?

Kevin-32B 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 Kevin-32B?

Its weights are published under the cognition-ai organisation on Hugging Face. We do not host model files — this site calculates what hardware is needed to run them.

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