Octo-Base TPS calculator

Open weights University of California (UC) Berkeley,Stanford University,Carnegie Mellon University (CMU),DeepMind 93M parameters May 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 · Q8_0 · 396 tok/s

Fastest card

B200

36,433 tok/s · 180 GB

Which GPUs can run Octo-Base?

Set the inputs, read the answer

A longer conversation needs more memory, which can push this model off smaller cards.

Hides cards that would only fit the model by compressing it below this point.

818 cards match

Calculating
Needs Quantisation Fit
36,433 tok/s

21,860–58,292 · low confidence

B200 NVIDIA 180 GB 8,000 GB/s Jan 2024 0.8 GB Q8_0 Comfortable
36,433 tok/s

21,860–58,292 · low confidence

B300 NVIDIA 288 GB 8,000 GB/s Sep 2025 0.8 GB Q8_0 Comfortable
29,092 tok/s

17,455–46,548 · low confidence

Radeon Instinct MI350X AMD 288 GB 8,190 GB/s Jan 2025 0.8 GB Q8_0 Comfortable
29,092 tok/s

17,455–46,548 · low confidence

Radeon Instinct MI355X AMD 288 GB 8,190 GB/s Jan 2025 0.8 GB Q8_0 Comfortable
23,267 tok/s

13,960–37,227 · low confidence

Radeon Instinct MI300 AMD 128 GB 6,550 GB/s Jan 2023 0.8 GB Q8_0 Comfortable
22,269 tok/s

13,362–35,631 · low confidence

H200 NVL NVIDIA 141 GB 4,890 GB/s Nov 2024 0.8 GB Q8_0 Comfortable
22,269 tok/s

13,362–35,631 · low confidence

H200 SXM 141 GB NVIDIA 141 GB 4,890 GB/s Nov 2024 0.8 GB Q8_0 Comfortable
21,313 tok/s

12,788–34,101 · low confidence

Radeon Instinct MI325X AMD 256 GB 6,000 GB/s Oct 2024 0.8 GB Q8_0 Comfortable
18,915 tok/s

11,349–30,265 · low confidence

Radeon Instinct MI300A AMD 128 GB 5,325 GB/s Dec 2023 0.8 GB Q8_0 Comfortable
18,915 tok/s

11,349–30,265 · low confidence

Radeon Instinct MI300X AMD 192 GB 5,325 GB/s Dec 2023 0.8 GB Q8_0 Comfortable
18,915 tok/s

11,349–30,265 · low confidence

Radeon Instinct MI308X AMD 192 GB 5,325 GB/s Dec 2023 0.8 GB Q8_0 Comfortable
17,943 tok/s

10,766–28,709 · low confidence

H100 NVL 94 GB NVIDIA 94 GB 3,940 GB/s Mar 2023 0.8 GB Q8_0 Comfortable
15,302 tok/s

9,181–24,483 · low confidence

H100 PCIe 96 GB NVIDIA 96 GB 3,360 GB/s Mar 2023 0.8 GB Q8_0 Comfortable
15,302 tok/s

9,181–24,483 · low confidence

H100 SXM5 80 GB NVIDIA 80 GB 3,360 GB/s Oct 2022 0.8 GB Q8_0 Comfortable
15,302 tok/s

9,181–24,483 · low confidence

H100 SXM5 94 GB NVIDIA 94 GB 3,360 GB/s Mar 2023 0.8 GB Q8_0 Comfortable
15,302 tok/s

9,181–24,483 · low confidence

H100 SXM5 96 GB NVIDIA 96 GB 3,360 GB/s Mar 2023 0.8 GB Q8_0 Comfortable
15,302 tok/s

9,181–24,483 · low confidence

H800 SXM5 NVIDIA 80 GB 3,360 GB/s Mar 2023 0.8 GB Q8_0 Comfortable
11,651 tok/s

6,991–18,642 · low confidence

Radeon Instinct MI250 AMD 128 GB 3,280 GB/s Nov 2021 0.8 GB Q8_0 Comfortable
11,651 tok/s

6,991–18,642 · low confidence

Radeon Instinct MI250X AMD 128 GB 3,280 GB/s Nov 2021 0.8 GB Q8_0 Comfortable
9,709 tok/s

5,826–15,535 · low confidence

Data Center GPU Max 1550 Intel 128 GB 3,280 GB/s Jan 2023 0.8 GB Q8_0 Comfortable
9,502 tok/s

5,701–15,203 · low confidence

Data Center GPU Max Subsystem Intel 128 GB 3,210 GB/s Jan 2023 0.8 GB Q8_0 Comfortable
9,290 tok/s

5,574–14,865 · low confidence

A100 SXM4 80 GB NVIDIA 80 GB 2,040 GB/s Nov 2020 0.8 GB Q8_0 Comfortable
9,290 tok/s

5,574–14,865 · low confidence

A100X NVIDIA 80 GB 2,040 GB/s Jun 2021 0.8 GB Q8_0 Comfortable
9,290 tok/s

5,574–14,865 · low confidence

A800 SXM4 80 GB NVIDIA 80 GB 2,040 GB/s Aug 2022 0.8 GB Q8_0 Comfortable
9,290 tok/s

5,574–14,865 · low confidence

H100 CNX NVIDIA 80 GB 2,040 GB/s Mar 2023 0.8 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
University of California (UC) Berkeley,Stanford University,Carnegie Mellon University (CMU),DeepMind
Organisation type
Academia,Academia,Academia,Industry
Country
United States of America, United Kingdom of Great Britain and Northern Ireland
Published
20 May 2024
Authors
Octo Model Team, Dibya Ghosh, Homer Walke, Karl Pertsch, Kevin Black, Oier Mees, Sudeep Dasari, Joey Hejna, Tobias Kreiman, Charles Xu, Jianlan Luo, You Liang Tan, Lawrence Yunliang Chen, Pannag Sanketi, Quan Vuong, Ted Xiao, Dorsa Sadigh, Chelsea Finn, Sergey Levine

What it does

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

Domain
Robotics
Task
Robotic manipulation
Numerical format
BF16

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
93M

Source: https://arxiv.org/abs/2405.12213

Training data
tokens

Octo was pre-trained on 800k robot episodes from the Open X-Embodiment dataset. The authors describe this dataset as "heterogeneous," so it's unclear how to compute the size of the dataset. (Source: https://arxiv.org/abs/2405.12213)

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
5.9 × 10²⁰ FLOP

Training compute = peak FLOPs * utilization rate * training time ~= 128 TPUs * 275e12 FLOPs / TPU * 0.33 * 14 hours * 3600 s / hour ~= 585446400e12 FLOPs = 5.85e20 FLOPs, assuming utilization rate = 0.33.

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
Google TPU v4
Chips used
128
Wall-clock time
14 hours

“We trained two variants of our model: Octo-Small with a transformer backbone that mirrors the size of a ViT-S, andOcto-Base with a transformer backbone that mirrors the size of a ViT-B. The ViT-B was trained for 300k steps with a batch size of 2048 using a TPU v4-128 pod, which took 14 hours." (Source: https://arxiv.org/abs/2405.12213)

Power draw
86.0 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 (unrestricted)
Training code
Open source

MIT license https://huggingface.co/rail-berkeley/octo-base https://github.com/octo-models/octo

Hugging Face
rail-berkeley

How it is classified

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

Why it is tracked
SOTA improvement

Octo’s ability to control multiple robots out-of-the-box to the best openly available generalist robot policy, RT-1-X. RT-1-X had success rates of 0.2, ~0.38, and 0.6 on WidowX, UR5, and RT-1 Robot. Octo-Base outperformed RT-1-X on all tasks, with success rates of approximately 0.5, 0.7, and 0.8 on WidowX, UR5, and RT-1 Robot, respectively. (Source: https://arxiv.org/abs/2405.12213)

Record confidence
Confident
Citations
1,244

Sources

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

Reference
Octo: An Open-Source Generalist Robot Policy
Last updated
25 May 2026

The extremes

What the numbers mean

What it takes to run this model

Minimum card

Tesla C1080

Memory needed

0.8 GB

Fastest

36,433 tok/s

Octo-Base is small enough at 93M 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 396 tokens per second.

Top of the range is the B200, at roughly 36,433 tokens per second thanks to 8,000 GB/s of bandwidth.

About this model

Octo-Base was published by University of California (UC) Berkeley,Stanford University,Carnegie Mellon University (CMU),DeepMind, in United States of America, in May 2024. The organisation is categorised as academia,Academia,Academia,Industry.

It works in Robotics, and is recorded as doing robotic manipulation.

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 rail-berkeley organisation on Hugging Face.

How fast it runs, and why

The median result is around 1,023.0 tokens per second; 818 cards produce text faster than most people read it.

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.

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.

How it was trained

The training run consumed about 5.9 × 10²⁰ FLOP, on Google TPU v4. That figure describes the cost of creating it and has no bearing on how quickly it generates text.

It is tracked in the underlying dataset for one reason in particular: sOTA improvement.

Step by step

How to choose a GPU for Octo-Base

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

  1. 01

    Check what it needs before anything else

    Every card here has been checked against Octo-Base — around 0.8 GB at Q8_0. Capacity is the gate — a card either holds it or it does not.

  2. 02

    Set the context length you will work at

    Longer conversations cost memory on top of what the weights need. Move the slider to your real working length before trusting any row for Octo-Base.

  3. 03

    Decide how much compression you will accept

    The quantisation column varies by card, because a bigger card holds a more accurate copy of Octo-Base — Q8_0 on the smallest card that fits. Set a floor to hold the comparison at one level.

  4. 04

    Compare tokens per second, not specifications

    The speed ordering for Octo-Base is effectively an ordering by memory bandwidth, which is why the B200 tops it at 36,433 tok/s.

  5. 05

    Read the fit column last

    The fit column separates cards that just manage Octo-Base from those with room to spare. Buy for the second if the context might grow.

  6. 06

    Open the card you have settled on

    Each card page repeats this sweep for every model we hold. It answers what else the hardware is good for, beyond Octo-Base.

Answers

Octo-Base — common questions

01

When was Octo-Base released?

Octo-Base was published in May 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.

02

What is Octo-Base used for?

Octo-Base works in Robotics, and is recorded as handling robotic manipulation. 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.

03

Where can I download Octo-Base?

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

04

How much compute was used to train Octo-Base?

Around 5.9 × 10²⁰ FLOP, on Google TPU v4. 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.

05

Can I run Octo-Base if it does not fit in my GPU?

It can be split between the card and system memory, but Octo-Base generates painfully slowly that way. Nothing on this page assumes offloading.

06

Would two GPUs run Octo-Base faster?

Two cards buy memory rather than speed. That matters for Octo-Base only if one card cannot hold it — 818 can, so a second adds little.

07

Why does the quantisation differ between cards for Octo-Base?

A larger card holds a more accurate copy. Across the cards that run Octo-Base, 1 compression levels are used; the floor control above pins it to one.

08

How accurate are these Octo-Base speed estimates?

These are estimates with real error bars. The fastest result here, 21,860–58,292 tok/s on the B200, could reasonably land anywhere in its published range depending on which runtime you use.

09

What GPU do I need to run Octo-Base?

The smallest card in our catalogue that holds Octo-Base is the Tesla C1080, with 4 GB of memory. It runs the model at Q8_0 using about 0.8 GB, and produces roughly 396 tokens per second. 818 cards in total can run it.

10

How fast is Octo-Base on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 36,433 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 Octo-Base clear that.

11

How much VRAM does Octo-Base need?

About 0.8 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.

12

Can I run Octo-Base on a 8 GB GPU?

Yes. A CMP 170HX 8 GB with 8 GB runs it at Q8_0, using about 0.8 GB and generating roughly 6,786 tokens per second — a comfortable fit.

13

Can I run Octo-Base on a 12 GB GPU?

Yes. A GeForce RTX 3080 Ti with 12 GB runs it at Q8_0, using about 0.8 GB and generating roughly 4,155 tokens per second — a comfortable fit.

14

Can I run Octo-Base on a 16 GB GPU?

Yes. A Tesla V100 SXM2 16 GB with 16 GB runs it at Q8_0, using about 0.8 GB and generating roughly 5,146 tokens per second — a comfortable fit.

15

Can I run Octo-Base on a 24 GB GPU?

Yes. A GeForce RTX 5090 D V2 with 24 GB runs it at Q8_0, using about 0.8 GB and generating roughly 6,102 tokens per second — a comfortable fit.

16

Is Octo-Base open source?

Its weights are published, so Octo-Base 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.

17

How many parameters does Octo-Base have?

Octo-Base has 93M parameters. Source: https://arxiv.org/abs/2405.12213. 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.

18

Who created Octo-Base?

Octo-Base was published by University of California (UC) Berkeley,Stanford University,Carnegie Mellon University (CMU),DeepMind, based in United States of America, categorised as academia,Academia,Academia,Industry.

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.