Earth-2 (cBottle-SR) TPS calculator

Open weights NVIDIA 330M 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

818 cards that can run it

818 cards we hold specifications for

Smallest card that fits

Tesla C1080

4 GB · Q8_0 · 112 tok/s

Fastest card

B200

10,267 tok/s · 180 GB

Which GPUs can run Earth-2 (cBottle-SR)?

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
10,267 tok/s 6,160–16,428 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	1.1 GB	Q8_0	Comfortable
10,267 tok/s 6,160–16,428 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	1.1 GB	Q8_0	Comfortable
8,199 tok/s 4,919–13,118 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	1.1 GB	Q8_0	Comfortable
8,199 tok/s 4,919–13,118 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	1.1 GB	Q8_0	Comfortable
6,557 tok/s 3,934–10,491 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	1.1 GB	Q8_0	Comfortable
6,276 tok/s 3,766–10,042 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	1.1 GB	Q8_0	Comfortable
6,276 tok/s 3,766–10,042 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	1.1 GB	Q8_0	Comfortable
6,006 tok/s 3,604–9,610 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	1.1 GB	Q8_0	Comfortable
5,331 tok/s 3,198–8,529 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	1.1 GB	Q8_0	Comfortable
5,331 tok/s 3,198–8,529 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	1.1 GB	Q8_0	Comfortable
5,331 tok/s 3,198–8,529 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	1.1 GB	Q8_0	Comfortable
5,057 tok/s 3,034–8,091 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	1.1 GB	Q8_0	Comfortable
4,312 tok/s 2,587–6,900 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	1.1 GB	Q8_0	Comfortable
4,312 tok/s 2,587–6,900 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	1.1 GB	Q8_0	Comfortable
4,312 tok/s 2,587–6,900 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	1.1 GB	Q8_0	Comfortable
4,312 tok/s 2,587–6,900 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	1.1 GB	Q8_0	Comfortable
4,312 tok/s 2,587–6,900 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	1.1 GB	Q8_0	Comfortable
3,284 tok/s 1,970–5,254 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	1.1 GB	Q8_0	Comfortable
3,284 tok/s 1,970–5,254 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	1.1 GB	Q8_0	Comfortable
2,736 tok/s 1,642–4,378 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	1.1 GB	Q8_0	Comfortable
2,678 tok/s 1,607–4,285 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	1.1 GB	Q8_0	Comfortable
2,618 tok/s 1,571–4,189 · low confidence	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	1.1 GB	Q8_0	Comfortable
2,618 tok/s 1,571–4,189 · low confidence	A100X NVIDIA	80 GB	2,040 GB/s	Jun 2021	1.1 GB	Q8_0	Comfortable
2,618 tok/s 1,571–4,189 · low confidence	A800 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Aug 2022	1.1 GB	Q8_0	Comfortable
2,618 tok/s 1,571–4,189 · low confidence	H100 CNX NVIDIA	80 GB	2,040 GB/s	Mar 2023	1.1 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: NVIDIA
Organisation type: Industry
Country: United States of America
Published: 10 May 2025
Authors: Noah D. Brenowitz, Tao Ge, Akshay Subramaniam, Aayush Gupta, David M. Hall, Morteza Mardani, Arash Vahdat, Karthik Kashinath, Michael S. Pritchard

What it does

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

Domain: Earth science, Image generation
Task: Image generation, Weather forecasting

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: 330M
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: 1.6 × 10²¹ 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
Chips used: 64
Chip-hours: 1,499
Power draw: 87.8 kW

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: Open source

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: Climate in a Bottle: Towards a Generative Foundation Model for the Kilometer-Scale Global Atmosphere
Last updated: 11 February 2026

The extremes

The ten fastest GPUs that run Earth-2 (cBottle-SR)

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 Earth-2 (cBottle-SR)

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

Tesla C1080

Memory needed

1.1 GB

Fastest

10,267 tok/s

Earth-2 (cBottle-SR) is small enough at 330M parameters that hardware is rarely the obstacle — 818 of the cards we track can run it, including cards several years old.

The smallest card that holds it is the Tesla C1080 with 4 GB, running it at Q8_0 and producing around 112 tokens per second.

At the other end, a B200 generates roughly 10,267 tokens per second on it, on the strength of 8,000 GB/s of memory bandwidth.

Where it came from

Earth-2 (cBottle-SR) was published by NVIDIA, in United States of America, in May 2025. The organisation is categorised as industry.

It works in Earth science, Image generation, and is recorded as doing image generation, Weather forecasting.

Because its weights were released, nothing about running it depends on a provider staying available — it is yours once downloaded.

Understanding the speeds

Half the cards that hold it manage more than 288.3 tokens per second, and 818 exceed reading speed outright.

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.

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.

Training and provenance

Training it took roughly 1.6 × 10²¹ FLOP of computation, on NVIDIA H100 SXM5 80GB — a measure of what producing the model cost, not of how fast it answers.

Step by step

How to choose a GPU for Earth-2 (cBottle-SR)

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

01

Check what it needs before anything else

The table lists every card that can hold Earth-2 (cBottle-SR) — around 1.1 GB at Q8_0. That figure, not the card's headline performance, is what decides whether it runs.
02

Decide how long your conversations run

The conversation occupies memory too, and grows as it goes. Set the slider to the length you expect: at long context Earth-2 (cBottle-SR) can slip off a card that handles short questions easily.
03

Choose how far you will compress it

The quantisation column varies by card, because a bigger card holds a more accurate copy of Earth-2 (cBottle-SR) — Q8_0 on the smallest card that fits. Set a floor to hold the comparison at one level.
04

Rank by throughput rather than spec sheet

Sort by speed to see how cards rank for Earth-2 (cBottle-SR). It will not match a gaming ordering — generation is bound by memory bandwidth, which is why the B200 tops it at 10,267 tok/s.
05

Check the fit verdict before buying

The fit column separates cards that just manage Earth-2 (cBottle-SR) from those with room to spare. Buy for the second if the context might grow.
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. Worth a look before buying for Earth-2 (cBottle-SR) alone — a card is usually bought for more than one model.

Answers

Earth-2 (cBottle-SR) — common questions

Would two GPUs run Earth-2 (cBottle-SR) faster?

Two cards buy memory rather than speed. That matters for Earth-2 (cBottle-SR) only if one card cannot hold it — 818 can, so a second adds little.

Why does the quantisation differ between cards for Earth-2 (cBottle-SR)?

A larger card holds a more accurate copy. Across the cards that run Earth-2 (cBottle-SR), 1 compression levels are used; the floor control above pins it to one.

How accurate are these Earth-2 (cBottle-SR) speed estimates?

They are calculated from specifications rather than measured, and each carries a range — 6,160–16,428 tok/s on the B200, for instance. The same model and card vary by thirty to fifty per cent depending on the inference software and its version.

What GPU do I need to run Earth-2 (cBottle-SR)?

The smallest card in our catalogue that holds Earth-2 (cBottle-SR) is the Tesla C1080, with 4 GB of memory. It runs the model at Q8_0 using about 1.1 GB, and produces roughly 112 tokens per second. 818 cards in total can run it.

How fast is Earth-2 (cBottle-SR) on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 10,267 tokens per second; the slowest that still runs it manages considerably less. Reading speed is around ten tokens per second, and 818 of the cards that can run Earth-2 (cBottle-SR) clear that.

How much VRAM does Earth-2 (cBottle-SR) need?

About 1.1 GB at Q8_0 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 Earth-2 (cBottle-SR) on a 8 GB GPU?

Yes. A CMP 170HX 8 GB with 8 GB runs it at Q8_0, using about 1.1 GB and generating roughly 1,912 tokens per second — a comfortable fit.

Can I run Earth-2 (cBottle-SR) on a 12 GB GPU?

Yes. A GeForce RTX 3080 Ti with 12 GB runs it at Q8_0, using about 1.1 GB and generating roughly 1,171 tokens per second — a comfortable fit.

Can I run Earth-2 (cBottle-SR) on a 16 GB GPU?

Yes. A Tesla V100 SXM2 16 GB with 16 GB runs it at Q8_0, using about 1.1 GB and generating roughly 1,450 tokens per second — a comfortable fit.

Can I run Earth-2 (cBottle-SR) on a 24 GB GPU?

Yes. A GeForce RTX 5090 D V2 with 24 GB runs it at Q8_0, using about 1.1 GB and generating roughly 1,720 tokens per second — a comfortable fit.

Is Earth-2 (cBottle-SR) open source?

Its weights are published, so Earth-2 (cBottle-SR) 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 Earth-2 (cBottle-SR) have?

Earth-2 (cBottle-SR) has 330M parameters. cBottle-SR has 330M 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.

Who created Earth-2 (cBottle-SR)?

Earth-2 (cBottle-SR) was published by NVIDIA, based in United States of America, categorised as industry.

When was Earth-2 (cBottle-SR) released?

Earth-2 (cBottle-SR) was published in May 2025.

What is Earth-2 (cBottle-SR) used for?

Earth-2 (cBottle-SR) works in Earth science, Image generation, and is recorded as handling image generation, Weather forecasting. 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 Earth-2 (cBottle-SR)?

The weights for Earth-2 (cBottle-SR) are published, though we do not hold a repository link for it. This site calculates hardware requirements rather than hosting model files.

How much compute was used to train Earth-2 (cBottle-SR)?

Around 1.6 × 10²¹ FLOP, on 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.

Can I run Earth-2 (cBottle-SR) 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 Earth-2 (cBottle-SR) is rarely worth using. Every figure here assumes the whole model is on the card.

Source

Original publication

Record last updated 11 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