Playground v2.5 TPS calculator

Open weights Playground 3.5B parameters February 2024

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 · Q5_K_M · 18.8 tok/s

Fastest card

B200

968 tok/s · 180 GB

Which GPUs can run Playground v2.5?

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
968 tok/s 581–1,549 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	4.4 GB	Q8_0	Comfortable
968 tok/s 581–1,549 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	4.4 GB	Q8_0	Comfortable
773 tok/s 464–1,237 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	4.4 GB	Q8_0	Comfortable
773 tok/s 464–1,237 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	4.4 GB	Q8_0	Comfortable
618 tok/s 371–989 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	4.4 GB	Q8_0	Comfortable
592 tok/s 355–947 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	4.4 GB	Q8_0	Comfortable
592 tok/s 355–947 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	4.4 GB	Q8_0	Comfortable
566 tok/s 340–906 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	4.4 GB	Q8_0	Comfortable
503 tok/s 302–804 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	4.4 GB	Q8_0	Comfortable
503 tok/s 302–804 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	4.4 GB	Q8_0	Comfortable
503 tok/s 302–804 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	4.4 GB	Q8_0	Comfortable
477 tok/s 286–763 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	4.4 GB	Q8_0	Comfortable
407 tok/s 244–651 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	4.4 GB	Q8_0	Comfortable
407 tok/s 244–651 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	4.4 GB	Q8_0	Comfortable
407 tok/s 244–651 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	4.4 GB	Q8_0	Comfortable
407 tok/s 244–651 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	4.4 GB	Q8_0	Comfortable
407 tok/s 244–651 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	4.4 GB	Q8_0	Comfortable
310 tok/s 186–495 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	4.4 GB	Q8_0	Comfortable
310 tok/s 186–495 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	4.4 GB	Q8_0	Comfortable
258 tok/s 155–413 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	4.4 GB	Q8_0	Comfortable
252 tok/s 151–404 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	4.4 GB	Q8_0	Comfortable
247 tok/s 148–395 · low confidence	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	4.4 GB	Q8_0	Comfortable
247 tok/s 148–395 · low confidence	A100X NVIDIA	80 GB	2,040 GB/s	Jun 2021	4.4 GB	Q8_0	Comfortable
247 tok/s 148–395 · low confidence	A800 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Aug 2022	4.4 GB	Q8_0	Comfortable
247 tok/s 148–395 · low confidence	H100 CNX NVIDIA	80 GB	2,040 GB/s	Mar 2023	4.4 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: Playground
Organisation type: Industry
Country: United States of America
Published: 27 February 2024
Authors: Daiqing Li, Aleks Kamko, Ehsan Akhgari, Ali Sabet, Linmiao Xu, Suhail Doshi

What it does

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

Domain: Image generation
Task: Text-to-image, Image 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: 3.5B
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 (restricted use)
Training code: Unreleased
Hugging Face: playgroundai

How it is classified

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

Record confidence: Unknown
Citations: 234

Sources

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

Reference: Playground v2.5: Three Insights towards Enhancing Aesthetic Quality in Text-to-Image Generation
Last updated: 25 May 2026

The extremes

The ten fastest GPUs that run Playground v2.5

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 Playground v2.5

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

3.2 GB

Fastest

968 tok/s

Playground v2.5 reaches a parameter count of 3.5B. 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: 818.

The least hardware that works is Tesla C1080, with a memory capacity of 4 GB, running it at a compression of Q5_K_M and producing around 18.8 tokens per second.

The fastest we calculate for it is B200, generating roughly 968 tokens per second on the strength of a memory bandwidth of 8,000 GB/s.

About this model

Playground v2.5 was published by Playground, in the country recorded as United States of America, during February 2024. The publishing organisation is categorised as industry.

It works in the domain of Image generation, and is recorded as performing the task of text-to-image, Image generation.

The weights being open is what puts this page in the calculator rather than only in the catalogue: it is a model you can actually hold. On Hugging Face it is published under the organisation playgroundai.

How fast it runs, and why

Across every card that can run it, the middle of the range sits at 32.5 tokens per second. Producing text faster than most people read it: 778 of them.

It is a dense model, so every parameter is read for every token produced. That makes speed track memory bandwidth almost exactly — a card with twice the bandwidth generates roughly twice as fast.

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.

Step by step

How to choose a GPU for Playground v2.5

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

The table lists every card able to hold Playground v2.5, needing around 3.2 GB at a compression of Q5_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 a card that seemed fine stops fitting Playground v2.5.
03

Decide how much compression you will accept

Each card runs the least-compressed copy it can hold, reaching a compression of Q5_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

Rank by throughput rather than spec sheet

Ranking by tokens per second follows memory bandwidth rather than core counts, for Playground v2.5. It will not match a gaming ordering, because generation is bound by memory bandwidth. The card topping the list is B200, at 968 tok/s.
05

Look at the headroom, not just the fit

The fit column separates cards that just manage it from those with room to spare, in the case of Playground v2.5. 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

Each card page repeats this sweep for every model we hold, answering what else the hardware is good for beyond Playground v2.5.

Answers

Playground v2.5 — common questions

Playground v2.5— 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. Every figure here assumes the whole model is resident on the card.

Playground v2.5— would two GPUs run it faster?

Capacity adds across cards; throughput does not. The number of cards already holding it on their own: 818. So a second card is rarely the answer here.

Playground v2.5— why does the quantisation differ between cards?

A larger card holds a more accurate copy. The number of compression levels used across the cards that run it: 2. Bigger cards get the more accurate version, and the quality floor above pins the comparison to one level.

Playground v2.5— how accurate are these speed estimates?

They are calculated from specifications rather than measured, and each carries a range. One example: 581–1,549 tok/s on B200. The same model and card vary by thirty to fifty per cent depending on the inference software and its version.

Playground v2.5— what GPU do I need to run it?

The smallest card in our catalogue that holds it is Tesla C1080, with a memory capacity of 4 GB. It runs the model at a compression of Q5_K_M using about 3.2 GB, and produces roughly 18.8 tokens per second. The number of cards able to run it in total: 818.

Playground v2.5— how fast is it on a GPU?

It depends on the card. The quickest we calculate is B200, at about 968 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: 778.

Playground v2.5— how much VRAM does it need?

It needs about 3.2 GB at a compression of Q5_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.

Playground v2.5— can I run it on a GPU holding 8 GB?

Yes. The card CMP 170HX 8 GB, holding 8 GB, runs it at a compression of Q8_0, using about 4.4 GB and generating roughly 180 tokens per second. The fit is comfortable.

Playground v2.5— 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 Q8_0, using about 4.4 GB and generating roughly 110 tokens per second. The fit is comfortable.

Playground v2.5— 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 Q8_0, using about 4.4 GB and generating roughly 137 tokens per second. The fit is comfortable.

Playground v2.5— 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 4.4 GB and generating roughly 162 tokens per second. The fit is comfortable.

Playground v2.5— 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.

Playground v2.5— how many parameters does it have?

It has a parameter count of 3.5B. The Model Description section of [1] says that the model “follows the same architecture as Stable Diffusion XL.” It also says that the model is a latent diffusion model that uses OpenCLIP-ViT/G and CLIP-ViT/L as text encoders, which seemingly would make it identical to Stable Diffusion XL [2]. Therefore, I will assume that Playground v2.5 has the same number of parameters as the Stable Diffusion XL 1.0 base model, which has 3.5B model parameters [3]. 1. https://huggingface.co/playgroundai/playground-v2.5-1024px-aesthetic 2. https://arxiv.org/pdf/2307.01952 3. https://stability.ai/news/stable-diffusion-sdxl-1-announcement. 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.

Playground v2.5— who created it?

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

Playground v2.5— when was it released?

It was published in February 2024. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.

Playground v2.5— what is it used for?

It works in the domain of Image generation, and is recorded as handling the task of text-to-image, Image generation. These are the areas it was designed around; they describe intent rather than a hard boundary.

Playground v2.5— where can I download it?

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

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