Calculate the TPS of the TITAN V CEO Edition on local AI models

NVIDIA 32 GB HBM2 868 GB/s June 2018

Every model in our catalogue assessed against this card at the context length and minimum quality you choose. Speed is an estimate for a single request, calculated from this card's memory bandwidth and the size of each model once compressed.

Calculated for this card

513 models it can run

679 models in our catalogue altogether

Largest model it holds

Phi-3.5-MoE

60.8B · Q3_K_M · 90.7 tok/s

Fastest model

Gemma 3 QAT 1B

368 tok/s · 1B

Which AI models can run on a TITAN V CEO Edition?

Set the inputs, read the answer

More context means more memory for the conversation cache. Speed is for a fresh conversation and does not change with this setting.

Hides models that would only fit by being compressed below this point.

513 models match

Calculating
Quantisation Fit
368 tok/s

313–441

Gemma 3 1B 1B Mar 2025 1.8 GB 33k tokens Q8_0 Comfortable
368 tok/s

313–441

Gemma 3 QAT 1B 1B Apr 2025 1.8 GB 33k tokens Q8_0 Comfortable
368 tok/s

221–588 · low confidence

HGRN 1B (WT 103) 1B Nov 2023 1.8 GB 131k tokens ? Q8_0 Comfortable
368 tok/s

221–588 · low confidence

LLama 3..2 Typhoon 2 1B 1B Dec 2024 1.8 GB 131k tokens ? Q8_0 Comfortable
368 tok/s

221–588 · low confidence

OLMo-1B 1B Feb 2024 1.8 GB 131k tokens ? Q8_0 Comfortable
368 tok/s

221–588 · low confidence

Pythia-1b 1B Apr 2023 1.8 GB 131k tokens ? Q8_0 Comfortable
341 tok/s

204–545 · low confidence

OpenELM-1.1B 1.1B May 2024 1.9 GB 131k tokens ? Q8_0 Comfortable
334 tok/s

201–535 · low confidence

DeciCoder-1B 1.1B Aug 2023 1.9 GB 131k tokens ? Q8_0 Comfortable
334 tok/s

201–535 · low confidence

SantaCoder 1.1B Jan 2023 1.9 GB 131k tokens ? Q8_0 Comfortable
334 tok/s

201–535 · low confidence

TinyLlama-1.1B (1T token checkpoint) 1.1B Oct 2023 1.9 GB 131k tokens ? Q8_0 Comfortable
334 tok/s

201–535 · low confidence

TinyLlama-1.1B (3T token checkpoint) 1.1B Oct 2023 1.9 GB 131k tokens ? Q8_0 Comfortable
306 tok/s

184–490 · low confidence

EXAONE 4.0 (1.2B) 1.2B Jul 2025 2.0 GB 131k tokens ? Q8_0 Comfortable
306 tok/s

184–490 · low confidence

MinerU2.5 1.2B Sep 2025 2.0 GB 131k tokens ? Q8_0 Comfortable
306 tok/s

184–490 · low confidence

Pleias 1.0 1.2B 1.2B Dec 2024 2.0 GB 131k tokens ? Q8_0 Comfortable
306 tok/s

184–490 · low confidence

Pleias-RAG-1B 1.2B Apr 2025 2.0 GB 131k tokens ? Q8_0 Comfortable
299 tok/s

254–359

Llama 3.2 1B 1.2B Sep 2024 2.2 GB 131k tokens Q8_0 Comfortable
295 tok/s

177–472 · low confidence

MiniCPM-1.2B 1.2B Jun 2024 2.0 GB 131k tokens ? Q8_0 Comfortable
283 tok/s

170–453 · low confidence

DeepSeek Coder 1.3B 1.3B Jan 2024 2.1 GB 131k tokens ? Q8_0 Comfortable
283 tok/s

170–453 · low confidence

DeepSeek-VL-1.3B 1.3B Mar 2024 2.1 GB 131k tokens ? Q8_0 Comfortable
283 tok/s

170–453 · low confidence

DigiRL 1.3B Jun 2024 2.1 GB 131k tokens ? Q8_0 Comfortable
283 tok/s

170–453 · low confidence

GLA Transformer 1.3B 1.3B Aug 2024 2.1 GB 131k tokens ? Q8_0 Comfortable
283 tok/s

170–453 · low confidence

Janus 1.3B 1.3B Oct 2024 2.1 GB 131k tokens ? Q8_0 Comfortable
283 tok/s

170–453 · low confidence

Kosmos-2.5 1.3B Aug 2024 2.1 GB 131k tokens ? Q8_0 Comfortable
283 tok/s

170–453 · low confidence

Otter 1.3B May 2023 2.1 GB 131k tokens ? Q8_0 Comfortable
283 tok/s

170–453 · low confidence

Phi-1 1.3B Oct 2023 2.1 GB 131k tokens ? 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

TITAN V CEO Edition full specification

Everything on record for this board, ordered by how much it bears on running a language model rather than by how a spec sheet would list it. Memory comes first because it decides the outcome; the rest is context.

Memory

The two specifications that decide what this card can run and how quickly. Capacity sets which models fit; bandwidth sets how many tokens per second they produce once they do.

Memory size
32 GB
Memory bandwidth
868 GB/s
Memory type
HBM2
Memory bus width
4,096 bit
Memory clock
848 MHz

The chip

Which processor is on the board and how it was manufactured. A smaller process size generally means more performance for the same power.

Graphics processor
GV100
Architecture
Volta
Generation
GeForce 10
Foundry
TSMC
Process size
12 nm
Transistors
21.1 billion
Transistor density
25,900 K/mm²
Die size
815 mm²
Released
21 June 2018

Clock speeds

How fast the processor runs. Worth far less here than on a gaming benchmark: generating text is limited by memory bandwidth, so a higher clock barely moves the result.

Base clock
1.2 GHz
Boost clock
1.46 GHz

Processing units

What the chip contains. These drive graphics performance and matter mainly for processing a long prompt rather than for producing the answer.

Shading units
5,120
Texture mapping units
320
Render output units
128
Streaming multiprocessors
80
Tensor cores
640
L1 cache
128 KB
L2 cache
6 MB

Theoretical performance

Peak arithmetic rates published for the board. These are ceilings that no real workload reaches, and generating text reaches a small fraction of them because it is limited by memory rather than arithmetic.

Half precision (FP16)
29.8 TFLOPS
Single precision (FP32)
14.9 TFLOPS
Double precision (FP64)
7.5 TFLOPS
Pixel rate
186 GPixel/s
Texture rate
466 GTexel/s

The board

What it takes to physically install and power the card — the practical constraints that decide whether it fits the machine you already own.

Power draw (TDP)
250 W
Suggested power supply
600 W
Power connectors
1x 6-pin + 1x 8-pin
Bus interface
PCIe 3.0 x16
Slot width
Dual-slot
Dimensions
267 mm × 40 mm
Display outputs
1x HDMI 2.0, 3x DisplayPort 1.4a

Software support

Which graphics and compute interfaces the card supports. CUDA compute capability is the one that bears on inference: below 7.0 there are no tensor cores, and modern inference software falls back to slower code paths.

CUDA compute capability
7.0
DirectX
12.1
OpenGL
4.6
Vulkan
1.4
OpenCL
3.0
Shader model
6.8

Listings

Where to buy a TITAN V CEO Edition

No vendor is currently listing this card. Listings come from vendors who publish them here directly — browse the vendor directory to see who is selling what.

What the numbers mean

Memory: the specification that decides everything

Memory

32 GB

Bandwidth

868 GB/s

Largest model

Phi-3.5-MoE

TITAN V CEO Edition carries 32 GB of HBM2. That covers the mid-sized models most people actually run. Once the runtime and driver reserve their working space, roughly this much is left: 28.8 GB.

Memory bandwidth reaches 868 GB/s across a bus of 4,096 bits. Generating a token means reading every weight once, so that figure sets the pace more than any other number here, and at this level text arrives faster than most people read.

Bandwidth is clock times bus width, and this card clocks its memory at 848 MHz. Widening the bus and raising the clock are the two levers a manufacturer has, which is why a card with unremarkable cores can still generate quickly.

Put together, the largest model that fits is Phi-3.5-MoE, 60.8B, compressed to Q3_K_M and generating around 90.7 tokens per second.

The chip and how it was built

TITAN V CEO Edition is built on the graphics processor GV100, using the architecture Volta from NVIDIA, as part of the generation GeForce 10.

The chip is manufactured by TSMC, on a process of 12 nm, with a die measuring 815 mm², holding 21.1 billion transistors. A smaller process generally means more performance for the same power, though for language models it matters far less than the memory subsystem.

It was released in June 2018, roughly 8.1120175294148 years ago. Inference software support tends to follow hardware by a year or two, so a card of this age generally has mature, well-optimised code paths available to it.

Compute throughput, and why it matters less than it looks

FP16

29.8 TFLOPS

FP64

7.5 TFLOPS

Tensor cores

640

On paper TITAN V CEO Edition reaches 29.8 TFLOPS at half precision, and 14.9 TFLOPS at single precision. These are peak figures no real workload sustains, and generating text reaches only a small fraction of them — decoding is limited by memory rather than arithmetic, which is why a card can look enormously powerful here and still produce tokens at an ordinary rate.

Double-precision throughput reaches 7.5 TFLOPS. It has no bearing on running a language model — no inference runtime uses it — but it separates datacentre parts from consumer ones, since the latter deliberately restrict it.

The card carries 640 tensor cores across 80 streaming multiprocessors. These accelerate the matrix arithmetic at the heart of a transformer, and they are what make prompt processing — reading a long document before answering — dramatically faster than it would otherwise be.

Clocks run from a base of 1.2 GHz to a boost of 1.46 GHz. Worth far less here than on a gaming benchmark: raising the clock speeds up the arithmetic, and the arithmetic is not what generation is waiting on.

Cache and processing units

TITAN V CEO Edition has an L1 cache of 128 KB, backed by an L2 cache of 6 MB. Cache absorbs a share of the memory traffic that would otherwise hit the main bus, which is the one place on this page where a number other than bandwidth quietly affects generation speed — a large L2 lets more of the working set stay close to the cores.

There are 5,120 shading units, 320 texture mapping units, and 128 render output units. These drive graphics workloads and contribute to prompt processing, but they sit idle for much of the time a model spends generating a reply.

Power, size and installation

Power draw

250 W

TITAN V CEO Edition is rated at 250 W, and the suggested system power supply is 600 W. Running a language model keeps a card busy in bursts rather than continuously — it draws hard while generating and idles between requests — so sustained draw over a working day is usually well below the rated figure.

The board occupies dual-slot, measuring 267 mm long, and needs 1x 6-pin + 1x 8-pin. Worth checking against the case and power supply already in the machine, since the largest cards need considerably more of both than a typical desktop provides.

It connects over PCIe 3.0 x16. The interface governs how quickly a model is loaded from disk into the card, not how fast it runs once there, so a narrower link costs a few seconds at startup and nothing thereafter.

The extremes

The largest AI models that run on a TITAN V CEO Edition

The biggest open-weight models that fit on this card, newest first. Each is shown at the best compression the card can hold.

  1. 01 Kimi Linear 48B · IQ4_XS · Oct 2025 18.8 tok/s
  2. 02 Llama Nemotron Super v1.5 49B · IQ4_XS · Jul 2025 18.4 tok/s
  3. 03 Nemotron-H 56B 56B · Q3_K_M · Apr 2025 17.7 tok/s
  4. 04 Nemotron-H 47B 47B · IQ4_XS · Apr 2025 19.2 tok/s
  5. 05 Llama Nemotron Super 49B 49B · IQ4_XS · Mar 2025 18.4 tok/s
  6. 06 Jamba 1.6 Mini 52B · IQ4_XS · Mar 2025 75.3 tok/s
  7. 07 Jamba 1.5 Mini 52B · IQ4_XS · Aug 2024 75.3 tok/s
  8. 08 Qwen2-57B-A14B 57B · IQ4_XS · Jun 2024 64.5 tok/s
  9. 09 Phi-3.5-MoE 60.8B · Q3_K_M · Apr 2024 90.7 tok/s
  10. 10 Jamba 51.6B · IQ4_XS · Mar 2024 75.3 tok/s

The fastest AI models on a TITAN V CEO Edition

Where this card produces tokens quickest. Smaller models dominate here, because generating each token means reading the whole model out of memory once.

  1. 01 Gemma 3 QAT 1B 1B · Q8_0 · 1.8 GB 368 tok/s
  2. 02 Gemma 3 1B 1B · Q8_0 · 1.8 GB 368 tok/s
  3. 03 LLama 3..2 Typhoon 2 1B 1B · Q8_0 · 1.8 GB 368 tok/s
  4. 04 OLMo-1B 1B · Q8_0 · 1.8 GB 368 tok/s
  5. 05 HGRN 1B (WT 103) 1B · Q8_0 · 1.8 GB 368 tok/s
  6. 06 Pythia-1b 1B · Q8_0 · 1.8 GB 368 tok/s
  7. 07 OpenELM-1.1B 1.1B · Q8_0 · 1.9 GB 341 tok/s
  8. 08 TinyLlama-1.1B (1T token checkpoint) 1.1B · Q8_0 · 1.9 GB 334 tok/s
  9. 09 TinyLlama-1.1B (3T token checkpoint) 1.1B · Q8_0 · 1.9 GB 334 tok/s
  10. 10 DeciCoder-1B 1.1B · Q8_0 · 1.9 GB 334 tok/s

Step by step

How to work out the tokens per second of a TITAN V CEO Edition

You do not have to calculate anything by hand — the gputps.com calculator on this page has already worked it out for every model this card can hold. Reading off the answer takes six steps.

  1. 01

    Find the model in the table

    The table lists 513 models this card runs. Search narrows the list by name or by size.

  2. 02

    Set the context length you will actually use

    Longer conversations cost memory on top of the weights. Against 32 GB it is often what pushes a large model over the edge.

  3. 03

    Choose how far you will compress

    By default the table picks the least-compressed copy that fits. Setting a floor removes models that only qualify through heavy compression.

  4. 04

    Take the range as the answer

    Each speed is an estimate for a single conversation, with a range beneath it. The top end here is 368 tok/s on Gemma 3 QAT 1B. The same card and model vary by thirty to fifty per cent between inference runtimes.

  5. 05

    Check the memory column before committing

    The fit column separates models that just fit from those with room to spare — worth checking before settling on one, against an available 32 GB.

  6. 06

    Open the model to compare cards

    Each model page repeats this calculation for the whole catalogue. Worth a look before deciding: it shows what else runs the same model, alongside TITAN V CEO Edition.

Answers

TITAN V CEO Edition — common questions

01

TITAN V CEO Edition— how much memory does it have?

This card has 32 GB of HBM2. Around a tenth is reserved by the inference runtime and the driver, leaving roughly 28.8 GB available for a model and its conversation.

02

TITAN V CEO Edition— what is its memory bandwidth?

Memory bandwidth reaches 868 GB/s across a bus of 4,096 bits. This is the single best predictor of how fast it generates text, because producing each token means reading the entire model out of memory once.

03

TITAN V CEO Edition— what type of memory does it use?

It uses HBM2 clocked at 848 MHz. HBM types are found on datacentre accelerators and carry far more bandwidth than the GDDR used on desktop cards, which is why they generate tokens considerably faster at the same capacity.

04

TITAN V CEO Edition— who makes it?

This is a product of NVIDIA, with the chip manufactured by TSMC, on a process of 12 nm.

05

TITAN V CEO Edition— when was it released?

It was released in June 2018.

06

TITAN V CEO Edition— how much power does it use?

Rated board power is 250 W, and the suggested system power supply is 600 W. Generating text draws hard in bursts and idles between requests, so average consumption over a working session is normally well below the rated figure.

07

TITAN V CEO Edition— how much cache does it have?

The L1 cache is 128 KB, and the L2 cache is 6 MB. Cache absorbs part of the memory traffic that would otherwise reach the main bus, so a larger L2 gives a modest lift to generation speed beyond what bandwidth alone predicts.

08

TITAN V CEO Edition— what are its TFLOPS?

It is rated at 29.8 TFLOPS at half precision and 14.9 TFLOPS at single precision. These are peak arithmetic ceilings rather than achievable rates, and text generation reaches only a small fraction of them because it is limited by memory bandwidth instead.

09

TITAN V CEO Edition— how many tensor cores does it have?

It has 640 tensor cores across 80 streaming multiprocessors. They accelerate the matrix arithmetic a transformer is built from, which mainly speeds up processing a long prompt rather than producing the reply.

10

TITAN V CEO Edition— does it support CUDA?

Yes. It reports CUDA compute capability 7.0. Capability 7.0 and above has tensor cores, which modern inference software uses; below that it falls back to slower code paths for quantised models.

11

TITAN V CEO Edition— what bus interface does it use?

It uses PCIe 3.0 x16. This governs how fast a model is loaded onto the card rather than how fast it runs once loaded, so it costs a few seconds at startup and nothing during generation.

12

TITAN V CEO Edition— is it good for running local AI models?

Its memory comfortably covers the mid-sized models most people run locally and its bandwidth is high enough to generate text faster than most people read. In total it runs 513 of the models we track. Whether that is enough depends entirely on which model you want — the table above answers that directly.

13

TITAN V CEO Edition— can it run a model that does not fit in its memory?

Offloading past the card's 32 GB is possible and usually a false economy: the system-memory portion is slow enough to dominate the result.

14

Would two TITAN V CEO Edition cards be twice as fast?

Capacity adds, throughput does not. Two of them give you 64 GB which lets you hold models neither could hold alone, but generation does not split that way. These figures describe one card.

15

TITAN V CEO Edition— which AI models can it run?

513 of the 679 open-weight language models we track fit on this card and can be run locally. The table on this page lists every one, with the memory it needs, the quantisation it runs at and an estimated generation speed.

16

TITAN V CEO Edition— what is the largest AI model it can run?

The largest model in our catalogue that fits is Phi-3.5-MoE at 60.8B parameters, compressed to Q3_K_M. It generates roughly 90.7 tokens per second and needs about 27.2 GB of the card's memory.

17

TITAN V CEO Edition— how many tokens per second does it produce?

It depends on the model. The fastest model we track here is Gemma 3 QAT 1B at about 368 tokens per second, while larger models run proportionally slower because each token requires reading the whole model out of memory once. Speeds are estimates for a single conversation at a time.

18

TITAN V CEO Edition— can it run 7B models?

Yes. For example it runs Multi-Token Prediction 7B at Q8_0, using about 7.9 GB of memory and generating around 54.9 tokens per second.

19

TITAN V CEO Edition— can it run 13B models?

Yes. For example it runs DeepSeekMoE-16B at Q8_0, using about 17.5 GB of memory and generating around 128 tokens per second.

20

TITAN V CEO Edition— can it run 30B models?

Yes. For example it runs ERNIE-4.5-VL-28B-A3B at Q6_K, using about 22.7 GB of memory and generating around 106 tokens per second.

The other direction

Looking at it from the other side?

This page starts from the hardware. If you already know which model you want and need to know what it takes to run it, start from the model instead.

All GPUs