Ovis1.6-Gemma2-27B TPS calculator

Open weights Alibaba 28.9B parameters November 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

132 cards that can run it

818 cards we hold specifications for

Smallest card that fits

RTX A4500

20 GB · IQ4_XS · 23.0 tok/s

Fastest card

B200

117 tok/s · 180 GB

Which GPUs can run Ovis1.6-Gemma2-27B?

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.

132 cards match

Calculating
Needs Quantisation Fit
117 tok/s

70–188 · low confidence

B200 NVIDIA 180 GB 8,000 GB/s Jan 2024 31.6 GB Q8_0 Comfortable
117 tok/s

70–188 · low confidence

B300 NVIDIA 288 GB 8,000 GB/s Sep 2025 31.6 GB Q8_0 Comfortable
93.6 tok/s

56–150 · low confidence

Radeon Instinct MI350X AMD 288 GB 8,190 GB/s Jan 2025 31.6 GB Q8_0 Comfortable
93.6 tok/s

56–150 · low confidence

Radeon Instinct MI355X AMD 288 GB 8,190 GB/s Jan 2025 31.6 GB Q8_0 Comfortable
74.9 tok/s

45–120 · low confidence

Radeon Instinct MI300 AMD 128 GB 6,550 GB/s Jan 2023 31.6 GB Q8_0 Comfortable
71.7 tok/s

43–115 · low confidence

H200 NVL NVIDIA 141 GB 4,890 GB/s Nov 2024 31.6 GB Q8_0 Comfortable
71.7 tok/s

43–115 · low confidence

H200 SXM 141 GB NVIDIA 141 GB 4,890 GB/s Nov 2024 31.6 GB Q8_0 Comfortable
68.6 tok/s

41–110 · low confidence

Radeon Instinct MI325X AMD 256 GB 6,000 GB/s Oct 2024 31.6 GB Q8_0 Comfortable
60.9 tok/s

37–97 · low confidence

Radeon Instinct MI300A AMD 128 GB 5,325 GB/s Dec 2023 31.6 GB Q8_0 Comfortable
60.9 tok/s

37–97 · low confidence

Radeon Instinct MI300X AMD 192 GB 5,325 GB/s Dec 2023 31.6 GB Q8_0 Comfortable
60.9 tok/s

37–97 · low confidence

Radeon Instinct MI308X AMD 192 GB 5,325 GB/s Dec 2023 31.6 GB Q8_0 Comfortable
57.7 tok/s

35–92 · low confidence

H100 NVL 94 GB NVIDIA 94 GB 3,940 GB/s Mar 2023 31.6 GB Q8_0 Comfortable
49.2 tok/s

30–79 · low confidence

H100 PCIe 96 GB NVIDIA 96 GB 3,360 GB/s Mar 2023 31.6 GB Q8_0 Comfortable
49.2 tok/s

30–79 · low confidence

H100 SXM5 80 GB NVIDIA 80 GB 3,360 GB/s Oct 2022 31.6 GB Q8_0 Comfortable
49.2 tok/s

30–79 · low confidence

H100 SXM5 94 GB NVIDIA 94 GB 3,360 GB/s Mar 2023 31.6 GB Q8_0 Comfortable
49.2 tok/s

30–79 · low confidence

H100 SXM5 96 GB NVIDIA 96 GB 3,360 GB/s Mar 2023 31.6 GB Q8_0 Comfortable
49.2 tok/s

30–79 · low confidence

H800 SXM5 NVIDIA 80 GB 3,360 GB/s Mar 2023 31.6 GB Q8_0 Comfortable
39.8 tok/s

24–64 · low confidence

DRIVE A100 PROD NVIDIA 32 GB 1,870 GB/s May 2020 24.9 GB Q6_K Tight
39.8 tok/s

24–64 · low confidence

GRID A100A NVIDIA 32 GB 1,870 GB/s May 2020 24.9 GB Q6_K Tight
38.1 tok/s

23–61 · low confidence

GeForce RTX 5090 NVIDIA 32 GB 1,790 GB/s Jan 2025 24.9 GB Q6_K Tight
38.1 tok/s

23–61 · low confidence

GeForce RTX 5090 D NVIDIA 32 GB 1,790 GB/s Jan 2025 24.9 GB Q6_K Tight
37.5 tok/s

23–60 · low confidence

Radeon Instinct MI250 AMD 128 GB 3,280 GB/s Nov 2021 31.6 GB Q8_0 Comfortable
37.5 tok/s

23–60 · low confidence

Radeon Instinct MI250X AMD 128 GB 3,280 GB/s Nov 2021 31.6 GB Q8_0 Comfortable
35.1 tok/s

21–56 · low confidence

GeForce RTX 5090 D V2 NVIDIA 24 GB 1,340 GB/s Aug 2025 21.5 GB Q5_K_M Tight
31.9 tok/s

19–51 · low confidence

A30X NVIDIA 24 GB 1,220 GB/s Apr 2021 21.5 GB Q5_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
Alibaba
Organisation type
Industry
Country
China
Published
26 November 2024

What it does

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

Domain
Multimodal, Language, Vision
Task
Language modeling/generation, Visual question answering
Base model
Gemma 2 27B,SigLIP 400M

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
28.9B
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 (unrestricted)
Hugging Face
AIDC-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
Ovis1.6-Gemma2-27B
Last updated
28 November 2025

The extremes

What the numbers mean

What it takes to run this model

Minimum card

RTX A4500

Memory needed

16.5 GB

Fastest

117 tok/s

With 28.9B parameters, Ovis1.6-Gemma2-27B 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 smallest card that holds it is the RTX A4500 with 20 GB, running it at IQ4_XS and producing around 23.0 tokens per second.

The quickest result comes from a B200 at around 117 tokens per second — its 8,000 GB/s of bandwidth is what buys that.

About this model

Ovis1.6-Gemma2-27B was published by Alibaba, in China, in November 2024. The organisation is categorised as industry.

It works in Multimodal, Language, Vision, and is recorded as doing language modeling/generation, Visual question answering.

It builds on Gemma 2 27B,SigLIP 400M, 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 AIDC-AI organisation on Hugging Face.

How fast it runs, and why

Across every card that can run it, the middle of the range is about 20.3 tokens per second, and 105 of them clear the ten tokens per second that roughly matches reading speed.

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.

Step by step

How to choose a GPU for Ovis1.6-Gemma2-27B

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

  1. 01

    Read the memory figure first

    The table lists every card that can hold Ovis1.6-Gemma2-27B — around 16.5 GB at IQ4_XS. That figure, not the card's headline performance, is what decides whether it runs.

  2. 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 Ovis1.6-Gemma2-27B can slip off a card that handles short questions easily.

  3. 03

    Decide how much compression you will accept

    Each card runs the least-compressed copy it can hold — IQ4_XS on the smallest card that fits. Setting a floor drops the cards that only manage Ovis1.6-Gemma2-27B by squeezing it further than you would want.

  4. 04

    Rank by throughput rather than spec sheet

    The speed ordering for Ovis1.6-Gemma2-27B is effectively an ordering by memory bandwidth, which is why the B200 tops it at 117 tok/s.

  5. 05

    Read the fit column last

    A tight fit runs Ovis1.6-Gemma2-27B but leaves nothing spare for a longer conversation; comfortable has headroom. If you expect to grow the context, buy for comfortable.

  6. 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 Ovis1.6-Gemma2-27B alone — a card is usually bought for more than one model.

Answers

Ovis1.6-Gemma2-27B — common questions

01

How much VRAM does Ovis1.6-Gemma2-27B need?

About 16.5 GB at IQ4_XS 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.

02

Can I run Ovis1.6-Gemma2-27B on a 24 GB GPU?

Yes. A GeForce RTX 5090 D V2 with 24 GB runs it at Q5_K_M, using about 21.5 GB and generating roughly 35.1 tokens per second — a tight fit.

03

Is Ovis1.6-Gemma2-27B open source?

Its weights are published, so Ovis1.6-Gemma2-27B 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.

04

How many parameters does Ovis1.6-Gemma2-27B have?

Ovis1.6-Gemma2-27B has 28.9B 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.

05

Who created Ovis1.6-Gemma2-27B?

Ovis1.6-Gemma2-27B was published by Alibaba, based in China, categorised as industry.

06

When was Ovis1.6-Gemma2-27B released?

Ovis1.6-Gemma2-27B was published in November 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.

07

What is Ovis1.6-Gemma2-27B used for?

Ovis1.6-Gemma2-27B works in Multimodal, Language, Vision, and is recorded as handling language modeling/generation, Visual question answering. 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.

08

Where can I download Ovis1.6-Gemma2-27B?

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

09

Can I run Ovis1.6-Gemma2-27B if it does not fit in my GPU?

Only by offloading, which is usually a false economy: the part in system memory drags the whole thing down — the nearest miss we calculate is short by 3.8 GB. Our figures for Ovis1.6-Gemma2-27B assume it is fully resident.

10

Would two GPUs run Ovis1.6-Gemma2-27B faster?

A second card roughly doubles the memory available but not the generation rate. With 132 cards already able to run Ovis1.6-Gemma2-27B alone, the case for pairing is weak.

11

Why does the quantisation differ between cards for Ovis1.6-Gemma2-27B?

A larger card holds a more accurate copy. Across the cards that run Ovis1.6-Gemma2-27B, 4 compression levels are used; the floor control above pins it to one.

12

How accurate are these Ovis1.6-Gemma2-27B 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 70–188 tok/s on the B200 rather than a single number.

13

What GPU do I need to run Ovis1.6-Gemma2-27B?

The smallest card in our catalogue that holds Ovis1.6-Gemma2-27B is the RTX A4500, with 20 GB of memory. It runs the model at IQ4_XS using about 16.5 GB, and produces roughly 23.0 tokens per second. 132 cards in total can run it.

14

How fast is Ovis1.6-Gemma2-27B on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 117 tokens per second; the slowest that still runs it manages considerably less. Reading speed is around ten tokens per second, and 105 of the cards that can run Ovis1.6-Gemma2-27B clear that.

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.