gpt-oss-20b TPS calculator

Open weights OpenAI 20.9B parameters August 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

241 cards that can run it

818 cards we hold specifications for

Smallest card that fits

Xeon Phi 7120P

16 GB · Q4_K_M · 10.7 tok/s

Fastest card

B200

162 tok/s · 180 GB

Which GPUs can run gpt-oss-20b?

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

241 cards match

Calculating

					Needs	Quantisation	Fit
162 tok/s 97–259 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	23.1 GB	Q8_0	Comfortable
162 tok/s 97–259 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	23.1 GB	Q8_0	Comfortable
129 tok/s 78–207 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	23.1 GB	Q8_0	Comfortable
129 tok/s 78–207 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	23.1 GB	Q8_0	Comfortable
103 tok/s 62–166 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	23.1 GB	Q8_0	Comfortable
99.1 tok/s 59–158 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	23.1 GB	Q8_0	Comfortable
99.1 tok/s 59–158 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	23.1 GB	Q8_0	Comfortable
94.8 tok/s 57–152 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	23.1 GB	Q8_0	Comfortable
84.1 tok/s 50–135 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	23.1 GB	Q8_0	Comfortable
84.1 tok/s 50–135 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	23.1 GB	Q8_0	Comfortable
84.1 tok/s 50–135 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	23.1 GB	Q8_0	Comfortable
79.8 tok/s 48–128 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	23.1 GB	Q8_0	Comfortable
68.1 tok/s 41–109 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	23.1 GB	Q8_0	Comfortable
68.1 tok/s 41–109 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	23.1 GB	Q8_0	Comfortable
68.1 tok/s 41–109 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	23.1 GB	Q8_0	Comfortable
68.1 tok/s 41–109 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	23.1 GB	Q8_0	Comfortable
68.1 tok/s 41–109 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	23.1 GB	Q8_0	Comfortable
52.8 tok/s 32–85 · low confidence	Tesla V100 SXM2 16 GB NVIDIA	16 GB	1,130 GB/s	Nov 2019	13.3 GB	Q4_K_M	Tight
51.8 tok/s 31–83 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	23.1 GB	Q8_0	Comfortable
51.8 tok/s 31–83 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	23.1 GB	Q8_0	Comfortable
44.9 tok/s 27–72 · low confidence	GeForce RTX 5080 NVIDIA	16 GB	960 GB/s	Jan 2025	13.3 GB	Q4_K_M	Tight
43.2 tok/s 26–69 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	23.1 GB	Q8_0	Comfortable
42.3 tok/s 25–68 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	23.1 GB	Q8_0	Comfortable
41.9 tok/s 25–67 · low confidence	Tesla V100 DGXS 16 GB NVIDIA	16 GB	897 GB/s	Mar 2018	13.3 GB	Q4_K_M	Tight
41.9 tok/s 25–67 · low confidence	Tesla V100 PCIe 16 GB NVIDIA	16 GB	897 GB/s	Jun 2017	13.3 GB	Q4_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: OpenAI
Organisation type: Industry
Country: United States of America
Published: 5 August 2025

What it does

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

Domain: Language
Task: Language modeling/generation, Question answering

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: 20.9B
Training data: tokens

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: 5.5 × 10²³ 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 H100 SXM5 80GB

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
Why it is tracked: Discretionary
Record confidence: Confident

Sources

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

Reference: gpt-oss-120b & gpt-oss-20b Model Card
Last updated: 10 February 2026

The extremes

The ten fastest GPUs that run gpt-oss-20b

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 gpt-oss-20b

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

Xeon Phi 7120P

Memory needed

13.3 GB

Fastest

162 tok/s

gpt-oss-20b reaches a parameter count of 20.9B. That lands in the range a serious desktop card can handle once the weights are compressed. The number of cards we track that can run it: 241.

The least hardware that works is Xeon Phi 7120P, with a memory capacity of 16 GB, running it at a compression of Q4_K_M and producing around 10.7 tokens per second.

The fastest we calculate for it is B200, generating roughly 162 tokens per second on the strength of a memory bandwidth of 8,000 GB/s.

Where it came from

gpt-oss-20b was published by OpenAI, in the country recorded as United States of America, during August 2025. The publishing organisation is categorised as industry.

It works in the domain of Language, and is recorded as performing the task of language modeling/generation, Question answering.

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 21.0 tokens per second. Exceeding reading speed outright: 198 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.

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.

What went into building it

Training it took a computation budget of roughly 5.5 × 10²³ FLOP, on hardware recorded as NVIDIA H100 SXM5 80GB. That figure measures what producing the model cost, and has no bearing on how fast it answers.

Its inclusion criterion: discretionary.

Step by step

How to choose a GPU for gpt-oss-20b

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 gpt-oss-20b, needing around 13.3 GB at a compression of Q4_K_M. That figure, not the headline performance of a card, is what decides whether it runs.
02

Match the context to your actual use

Longer conversations cost memory on top of what the weights need. Move the slider to your real working length before trusting any row for gpt-oss-20b.
03

Decide how much compression you will accept

Each card runs the least-compressed copy it can hold, reaching a compression of Q4_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

Ranking by tokens per second follows memory bandwidth rather than core counts, for gpt-oss-20b. It will not match a gaming ordering, because generation is bound by memory bandwidth. The card topping the list is B200, at 162 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 gpt-oss-20b. 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

See what else that card runs

Each card page repeats this sweep for every model we hold, answering what else the hardware is good for beyond gpt-oss-20b.

Answers

gpt-oss-20b — common questions

gpt-oss-20b— how many parameters does it have?

It has a parameter count of 20.9B. Total parameters: 20.91B. 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.

gpt-oss-20b— who created it?

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

gpt-oss-20b— when was it released?

It was published in August 2025.

gpt-oss-20b— what is it used for?

It works in the domain of Language, and is recorded as handling the task of language modeling/generation, Question answering. These are the areas it was designed around; they describe intent rather than a hard boundary.

gpt-oss-20b— 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.

gpt-oss-20b— how much compute was used to train it?

Training consumed around 5.5 × 10²³ FLOP, on hardware recorded as NVIDIA H100 SXM5 80GB. 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.

gpt-oss-20b— can I run it if it does not fit in my GPU?

It can be split between the card and system memory, but it generates painfully slowly that way. The nearest miss we calculate falls short by 2.5 GB. Every figure here assumes the whole model is resident on the card.

gpt-oss-20b— would two GPUs run it faster?

A second card roughly doubles the memory available but not the generation rate. The number already able to run it alone: 241. So a second card is rarely the answer here.

gpt-oss-20b— why does the quantisation differ between cards?

Each card is shown running the least-compressed copy it can hold. The number of distinct compression levels across the cards that fit it: 4. Bigger cards get the more accurate version, and the quality floor above pins the comparison to one level.

gpt-oss-20b— how accurate are these speed estimates?

They are calculated from specifications rather than measured, and each carries a range. One example: 97–259 tok/s on B200. The same model and card vary by thirty to fifty per cent depending on the inference software and its version.

gpt-oss-20b— what GPU do I need to run it?

The smallest card in our catalogue that holds it is Xeon Phi 7120P, with a memory capacity of 16 GB. It runs the model at a compression of Q4_K_M using about 13.3 GB, and produces roughly 10.7 tokens per second. The number of cards able to run it in total: 241.

gpt-oss-20b— how fast is it on a GPU?

It depends on the card. The quickest we calculate is B200, at about 162 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: 198.

gpt-oss-20b— how much VRAM does it need?

It needs about 13.3 GB at a compression of Q4_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.

gpt-oss-20b— 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 Q4_K_M, using about 13.3 GB and generating roughly 52.8 tokens per second. The fit is tight.

gpt-oss-20b— 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 Q6_K, using about 18.2 GB and generating roughly 39.4 tokens per second. The fit is tight.

gpt-oss-20b— 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.

Source

Original publication

Record last updated 10 February 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