InternVL2_5-26B TPS calculator

Open weights Shanghai AI Lab,SenseTime,Tsinghua University,Nanjing University,Fudan University,Chinese University of Hong Kong (CUHK),Shanghai Jiao Tong University 25.5B parameters December 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

241 cards that can run it

818 cards we hold specifications for

Smallest card that fits

Xeon Phi 7120P

16 GB · Q3_K_M · 10.3 tok/s

Fastest card

B200

133 tok/s · 180 GB

Which GPUs can run InternVL2_5-26B?

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.

241 cards match

Calculating
Needs Quantisation Fit
133 tok/s

80–213 · low confidence

B200 NVIDIA 180 GB 8,000 GB/s Jan 2024 28.0 GB Q8_0 Comfortable
133 tok/s

80–213 · low confidence

B300 NVIDIA 288 GB 8,000 GB/s Sep 2025 28.0 GB Q8_0 Comfortable
106 tok/s

64–170 · low confidence

Radeon Instinct MI350X AMD 288 GB 8,190 GB/s Jan 2025 28.0 GB Q8_0 Comfortable
106 tok/s

64–170 · low confidence

Radeon Instinct MI355X AMD 288 GB 8,190 GB/s Jan 2025 28.0 GB Q8_0 Comfortable
84.9 tok/s

51–136 · low confidence

Radeon Instinct MI300 AMD 128 GB 6,550 GB/s Jan 2023 28.0 GB Q8_0 Comfortable
81.2 tok/s

49–130 · low confidence

H200 NVL NVIDIA 141 GB 4,890 GB/s Nov 2024 28.0 GB Q8_0 Comfortable
81.2 tok/s

49–130 · low confidence

H200 SXM 141 GB NVIDIA 141 GB 4,890 GB/s Nov 2024 28.0 GB Q8_0 Comfortable
77.7 tok/s

47–124 · low confidence

Radeon Instinct MI325X AMD 256 GB 6,000 GB/s Oct 2024 28.0 GB Q8_0 Comfortable
69.0 tok/s

41–110 · low confidence

Radeon Instinct MI300A AMD 128 GB 5,325 GB/s Dec 2023 28.0 GB Q8_0 Comfortable
69.0 tok/s

41–110 · low confidence

Radeon Instinct MI300X AMD 192 GB 5,325 GB/s Dec 2023 28.0 GB Q8_0 Comfortable
69.0 tok/s

41–110 · low confidence

Radeon Instinct MI308X AMD 192 GB 5,325 GB/s Dec 2023 28.0 GB Q8_0 Comfortable
65.4 tok/s

39–105 · low confidence

H100 NVL 94 GB NVIDIA 94 GB 3,940 GB/s Mar 2023 28.0 GB Q8_0 Comfortable
55.8 tok/s

33–89 · low confidence

H100 PCIe 96 GB NVIDIA 96 GB 3,360 GB/s Mar 2023 28.0 GB Q8_0 Comfortable
55.8 tok/s

33–89 · low confidence

H100 SXM5 80 GB NVIDIA 80 GB 3,360 GB/s Oct 2022 28.0 GB Q8_0 Comfortable
55.8 tok/s

33–89 · low confidence

H100 SXM5 94 GB NVIDIA 94 GB 3,360 GB/s Mar 2023 28.0 GB Q8_0 Comfortable
55.8 tok/s

33–89 · low confidence

H100 SXM5 96 GB NVIDIA 96 GB 3,360 GB/s Mar 2023 28.0 GB Q8_0 Comfortable
55.8 tok/s

33–89 · low confidence

H800 SXM5 NVIDIA 80 GB 3,360 GB/s Mar 2023 28.0 GB Q8_0 Comfortable
50.6 tok/s

30–81 · low confidence

Tesla V100 SXM2 16 GB NVIDIA 16 GB 1,130 GB/s Nov 2019 13.2 GB Q3_K_M Tight
43.0 tok/s

26–69 · low confidence

GeForce RTX 5080 NVIDIA 16 GB 960 GB/s Jan 2025 13.2 GB Q3_K_M Tight
42.5 tok/s

26–68 · low confidence

Radeon Instinct MI250 AMD 128 GB 3,280 GB/s Nov 2021 28.0 GB Q8_0 Comfortable
42.5 tok/s

26–68 · low confidence

Radeon Instinct MI250X AMD 128 GB 3,280 GB/s Nov 2021 28.0 GB Q8_0 Comfortable
40.2 tok/s

24–64 · low confidence

Tesla V100 DGXS 16 GB NVIDIA 16 GB 897 GB/s Mar 2018 13.2 GB Q3_K_M Tight
40.2 tok/s

24–64 · low confidence

Tesla V100 PCIe 16 GB NVIDIA 16 GB 897 GB/s Jun 2017 13.2 GB Q3_K_M Tight
40.2 tok/s

24–64 · low confidence

Tesla V100 SXM2 16 GB NVIDIA 16 GB 897 GB/s Jun 2017 13.2 GB Q3_K_M Tight
40.2 tok/s

24–64 · low confidence

GeForce RTX 5070 Ti NVIDIA 16 GB 896 GB/s Feb 2025 13.2 GB Q3_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
Shanghai AI Lab,SenseTime,Tsinghua University,Nanjing University,Fudan University,Chinese University of Hong Kong (CUHK),Shanghai Jiao Tong University
Organisation type
Academia,Industry,Academia,Academia,Academia,Academia,Academia
Country
China, Hong Kong
Published
6 December 2024
Authors
Zhe Chen, Weiyun Wang, Yue Cao, Yangzhou Liu, Zhangwei Gao, Erfei Cui, Jinguo Zhu, Shenglong Ye, Hao Tian, Zhaoyang Liu, Lixin Gu, Xuehui Wang, Qingyun Li, Yimin Ren, Zixuan Chen, Jiapeng Luo, Jiahao Wang, Tan Jiang, Bo Wang, Conghui He, Botian Shi, Xingcheng Zhang, Han Lv, Yi Wang, Wenqi Shao, Pei Chu, Zhongying Tu, Tong He, Zhiyong Wu, Huipeng Deng, Jiaye Ge, Kai Chen, Kaipeng Zhang, Limin Wang,…

What it does

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

Domain
Multimodal, Language, Vision
Task
Visual question answering, Language modeling/generation
Base model
InternLM2.5,InternViT-6B

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
25.5B

25.5B

Training data
221,000,000,000 tokens

∼221B tokens

Epochs
1

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.

How it was established
Operation counting
Fine-tuning compute
1.2 × 10²² FLOP

Stage 1: 31B tokens (MLP projector 116.43M parameters + https://internvl.github.io/blog/2024-12-05-InternVL-2.5) Stage 2: 146B tokens (ViT 5.54B + MLP projector 116.43M) Stage 3: 44B tokens (full model) 6 FLOP / parameter / token * (116.43 * 10^6 parameters * 31 * 10^6 tokens + 5.656 * 10^9 parameters * 146 * 10^9 tokens + 26 * 10^9 parameters * 44 * 10^9 tokens) = 1.1818678e+22 FLOP

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://github.com/OpenGVLab/InternVL "Release training / evaluation code for InternVL2.5 series Support liger kernels to save GPU memory" is checked off https://huggingface.co/OpenGVLab/InternVL2_5-26B

Hugging Face
OpenGVLab

How it is classified

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

Record confidence
Confident
Citations
1,468

Sources

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

Reference
Expanding Performance Boundaries of Open-Source Multimodal Models with Model, Data, and Test-Time Scaling
Last updated
25 May 2026

The extremes

The ten fastest GPUs that run InternVL2_5-26B

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.

  1. 01 B300 288 GB · 8,000 GB/s · Q8_0 133 tok/s
  2. 02 B200 180 GB · 8,000 GB/s · Q8_0 133 tok/s
  3. 03 Radeon Instinct MI350X 288 GB · 8,190 GB/s · Q8_0 106 tok/s
  4. 04 Radeon Instinct MI355X 288 GB · 8,190 GB/s · Q8_0 106 tok/s
  5. 05 Radeon Instinct MI300 128 GB · 6,550 GB/s · Q8_0 84.9 tok/s
  6. 06 H200 NVL 141 GB · 4,890 GB/s · Q8_0 81.2 tok/s
  7. 07 H200 SXM 141 GB 141 GB · 4,890 GB/s · Q8_0 81.2 tok/s
  8. 08 Radeon Instinct MI325X 256 GB · 6,000 GB/s · Q8_0 77.7 tok/s
  9. 09 Radeon Instinct MI300A 128 GB · 5,325 GB/s · Q8_0 69.0 tok/s
  10. 10 Radeon Instinct MI300X 192 GB · 5,325 GB/s · Q8_0 69.0 tok/s

The smallest GPUs that still run InternVL2_5-26B

The cheapest route in, by memory capacity. A tight fit runs the model but leaves nothing spare for a longer conversation.

  1. 01 Ryzen AI Z2 Extreme GPU 16 GB · needs 13.2 GB · Q3_K_M · tight 9.0 tok/s
  2. 02 Radeon RX 7700 16 GB · needs 13.2 GB · Q3_K_M · tight 21.8 tok/s
  3. 03 Arc Pro B50 16 GB · needs 13.2 GB · Q3_K_M · tight 6.5 tok/s
  4. 04 RTX PRO 2000 Blackwell 16 GB · needs 13.2 GB · Q3_K_M · tight 12.9 tok/s
  5. 05 Ryzen Z2 Extreme GPU 16 GB · needs 13.2 GB · Q3_K_M · tight 4.5 tok/s
  6. 06 Radeon RX 9060 XT 16 GB 16 GB · needs 13.2 GB · Q3_K_M · tight 11.3 tok/s
  7. 07 GeForce RTX 5060 Ti 16 GB 16 GB · needs 13.2 GB · Q3_K_M · tight 20.1 tok/s
  8. 08 GeForce RTX 5080 Mobile 16 GB · needs 13.2 GB · Q3_K_M · tight 40.2 tok/s
  9. 09 Radeon RX 9070 16 GB · needs 13.2 GB · Q3_K_M · tight 22.5 tok/s
  10. 10 Radeon RX 9070 XT 16 GB · needs 13.2 GB · Q3_K_M · tight 22.5 tok/s

What the numbers mean

The hardware side

Minimum card

Xeon Phi 7120P

Memory needed

13.2 GB

Fastest

133 tok/s

With 25.5B parameters, InternVL2_5-26B lands in the range a serious desktop card can handle once the weights are compressed. 241 of the cards we track can run it.

At the low end, a Xeon Phi 7120P handles it — 16 GB, at Q3_K_M, for about 10.3 tokens per second.

At the other end, a B200 generates roughly 133 tokens per second on it, on the strength of 8,000 GB/s of memory bandwidth.

About this model

InternVL2_5-26B was published by Shanghai AI Lab,SenseTime,Tsinghua University,Nanjing University,Fudan University,Chinese University of Hong Kong (CUHK),Shanghai Jiao Tong University, in China, in December 2024. The organisation is categorised as academia,Industry,Academia,Academia,Academia,Academia,Academia.

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

It is derived from InternLM2.5,InternViT-6B rather than trained from scratch, which is the usual way a specialised model is produced.

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. It is published under the OpenGVLab organisation on Hugging Face.

How fast it runs, and why

The median result is around 19.9 tokens per second; 195 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.

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

The training set ran to roughly 221,000,000,000 tokens.

Step by step

How to choose a GPU for InternVL2_5-26B

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 InternVL2_5-26B — around 13.2 GB at Q3_K_M. 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 InternVL2_5-26B.

  3. 03

    Set a quality floor

    Compression is what makes InternVL2_5-26B fit smaller cards, at some cost in accuracy — Q3_K_M on the smallest card that fits. A minimum quality removes the ones that go too far.

  4. 04

    Rank by throughput rather than spec sheet

    Ranking by tokens per second for InternVL2_5-26B follows memory bandwidth, not core counts, which is why the B200 tops it at 133 tok/s.

  5. 05

    Check the fit verdict before buying

    A tight fit runs InternVL2_5-26B but leaves nothing spare for a longer conversation; comfortable has headroom. If you expect to grow the context, buy for comfortable.

  6. 06

    Check the card from the other side

    Every card name links to its own page, which runs the same calculation across the whole model catalogue. Worth a look before buying for InternVL2_5-26B alone — a card is usually bought for more than one model.

Answers

InternVL2_5-26B — common questions

01

Why does the quantisation differ between cards for InternVL2_5-26B?

Each card is shown running the least-compressed copy it can hold, and InternVL2_5-26B appears at 5 different compression levels across the cards that fit it. Bigger cards get the more accurate version.

02

How accurate are these InternVL2_5-26B speed estimates?

They are calculated from specifications rather than measured, and each carries a range — 80–213 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.

03

What GPU do I need to run InternVL2_5-26B?

The smallest card in our catalogue that holds InternVL2_5-26B is the Xeon Phi 7120P, with 16 GB of memory. It runs the model at Q3_K_M using about 13.2 GB, and produces roughly 10.3 tokens per second. 241 cards in total can run it.

04

How fast is InternVL2_5-26B on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 133 tokens per second; the slowest that still runs it manages considerably less. Reading speed is around ten tokens per second, and 195 of the cards that can run InternVL2_5-26B clear that.

05

How much VRAM does InternVL2_5-26B need?

About 13.2 GB at Q3_K_M 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.

06

Can I run InternVL2_5-26B on a 16 GB GPU?

Yes. A Tesla V100 SXM2 16 GB with 16 GB runs it at Q3_K_M, using about 13.2 GB and generating roughly 50.6 tokens per second — a tight fit.

07

Can I run InternVL2_5-26B on a 24 GB GPU?

Yes. A GeForce RTX 5090 D V2 with 24 GB runs it at Q5_K_M, using about 19.1 GB and generating roughly 39.8 tokens per second — a tight fit.

08

Is InternVL2_5-26B open source?

Its weights are published, so InternVL2_5-26B 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.

09

How many parameters does InternVL2_5-26B have?

InternVL2_5-26B has 25.5B parameters. 25.5B. 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.

10

Who created InternVL2_5-26B?

InternVL2_5-26B was published by Shanghai AI Lab,SenseTime,Tsinghua University,Nanjing University,Fudan University,Chinese University of Hong Kong (CUHK),Shanghai Jiao Tong University, based in China, categorised as academia,Industry,Academia,Academia,Academia,Academia,Academia.

11

When was InternVL2_5-26B released?

InternVL2_5-26B was published in December 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.

12

What is InternVL2_5-26B used for?

InternVL2_5-26B works in Multimodal, Language, Vision, and is recorded as handling visual question answering, Language modeling/generation. 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.

13

Where can I download InternVL2_5-26B?

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

14

Can I run InternVL2_5-26B 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 InternVL2_5-26B is rarely worth using — the nearest miss we calculate is short by 5.3 GB. Every figure here assumes the whole model is on the card.

15

Would two GPUs run InternVL2_5-26B faster?

Capacity adds across cards; throughput does not. Since 241 of the cards we track already hold InternVL2_5-26B on their own, a second card is rarely the answer here.

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