Calculate the TPS of the RTX PRO 6000 Blackwell Server on local AI models

NVIDIA 96 GB GDDR7 1,790 GB/s March 2025

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

609 of 679 models it can run

Largest model it holds

dots.llm1

142B · IQ4_XS · 13.1 tok/s

Fastest model

Gemma 3 QAT 1B

758 tok/s · 1B

What AI models can a RTX PRO 6000 Blackwell Server run?

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.

609 models match

Calculating
Quantisation Fit
758 tok/s

644–910

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

644–910

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

455–1,213 · low confidence

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

455–1,213 · low confidence

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

455–1,213 · low confidence

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

455–1,213 · low confidence

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

421–1,123 · low confidence

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

414–1,103 · low confidence

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

414–1,103 · low confidence

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

414–1,103 · low confidence

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

414–1,103 · low confidence

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

379–1,011 · low confidence

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

379–1,011 · low confidence

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

379–1,011 · low confidence

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

379–1,011 · low confidence

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

524–740

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

365–972 · low confidence

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

350–933 · low confidence

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

350–933 · low confidence

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

350–933 · low confidence

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

350–933 · low confidence

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

350–933 · low confidence

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

350–933 · low confidence

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

350–933 · low confidence

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

350–933 · 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

RTX PRO 6000 Blackwell Server 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
96 GB
Memory bandwidth
1,790 GB/s
Memory type
GDDR7
Memory bus width
512 bit
Memory clock
1.75 GHz

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
GB202
Architecture
Blackwell 2.0
Generation
Server Blackwell(Bxx)
Foundry
TSMC
Process size
5 nm
Transistors
92.2 billion
Transistor density
122,900 K/mm²
Die size
750 mm²
Released
18 March 2025

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
1.59 GHz
Boost clock
2.62 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
24,064
Texture mapping units
752
Render output units
192
Streaming multiprocessors
188
Tensor cores
752
Ray tracing cores
188
L1 cache
128 KB
L2 cache
128 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)
126 TFLOPS
Single precision (FP32)
126 TFLOPS
Double precision (FP64)
2 TFLOPS
Pixel rate
503 GPixel/s
Texture rate
1,968 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)
600 W
Suggested power supply
1,000 W
Power connectors
1x 16-pin
Bus interface
PCIe 5.0 x16
Slot width
Dual-slot
Dimensions
267 mm × 40 mm
Display outputs
4x DisplayPort 2.1b

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
12.0
DirectX
12.2
OpenGL
4.6
Vulkan
1.4
OpenCL
3.0
Shader model
6.8

Listings

Where to buy a RTX PRO 6000 Blackwell Server

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

Why memory is the number that matters here

Memory

96 GB

Bandwidth

1,790 GB/s

Largest model

dots.llm1

With 96 GB of GDDR7, the RTX PRO 6000 Blackwell Server is in the class of hardware that holds the largest open-weight models without splitting them across machines. Roughly 86.4 GB of that is reachable by an inference runtime once the driver takes its share.

Bandwidth is 1,790 GB/s across a 512-bit bus. 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.

That comes from a 1.75 GHz memory clock across the bus width above. 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.

In practice that combination tops out at dots.llm1 — 142B, compressed to IQ4_XS, generating around 13.1 tokens per second.

The chip and how it was built

The RTX PRO 6000 Blackwell Server is built on the GB202 graphics processor, using NVIDIA's Blackwell 2.0 architecture, as part of the Server Blackwell(Bxx) generation.

The chip is manufactured by TSMC, on a 5 nm process, with a die measuring 750 mm², holding 92.2 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 March 2025, roughly 1 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

126 TFLOPS

FP64

2 TFLOPS

Tensor cores

752

On paper the RTX PRO 6000 Blackwell Server reaches 126 TFLOPS at half precision and 126 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 is 2 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 752 tensor cores across 188 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 1.59 GHz at base to 2.62 GHz boosted. 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

The RTX PRO 6000 Blackwell Server has 128 KB of L1 cache, backed by 128 MB of L2. 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 24,064 shading units, 752 texture mapping units, and 192 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

600 W

The RTX PRO 6000 Blackwell Server is rated at 600 W, with a 1,000 W power supply suggested for the whole system. 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 a dual-slot, measuring 267 mm long, and needs 1x 16-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 5.0 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 a RTX PRO 6000 Blackwell Server can run

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 Mistral Medium 3.5 128B · Q4_K_M · Apr 2026 13.7 tok/s
  2. 02 dots.llm1 142B · IQ4_XS · Jul 2025 13.1 tok/s
  3. 03 Pixtral Large 124B · Q4_K_M · Nov 2024 14.1 tok/s
  4. 04 xLAM-8x22B 141B · Q4_K_M · Sep 2024 12.4 tok/s
  5. 05 SaulLM-large 141B · Q4_K_M · Jul 2024 12.4 tok/s
  6. 06 Mixtral 8x22B 141B · Q4_K_M · Apr 2024 44.9 tok/s
  7. 07 WizardLM-2 8x22B 141B · Q4_K_M · Apr 2024 12.4 tok/s
  8. 08 Zephyr 141B-A39B 141B · Q4_K_M · Apr 2024 44.9 tok/s
  9. 09 APUS-xDAN-4.0(MoE) 136B · Q4_K_M · Apr 2024 12.9 tok/s
  10. 10 DBRX 132B · Q4_K_M · Mar 2024 48.6 tok/s

The fastest AI models on a RTX PRO 6000 Blackwell Server

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 758 tok/s
  2. 02 Gemma 3 1B 1B · Q8_0 · 1.8 GB 758 tok/s
  3. 03 LLama 3..2 Typhoon 2 1B 1B · Q8_0 · 1.8 GB 758 tok/s
  4. 04 OLMo-1B 1B · Q8_0 · 1.8 GB 758 tok/s
  5. 05 HGRN 1B (WT 103) 1B · Q8_0 · 1.8 GB 758 tok/s
  6. 06 Pythia-1b 1B · Q8_0 · 1.8 GB 758 tok/s
  7. 07 OpenELM-1.1B 1.1B · Q8_0 · 1.9 GB 702 tok/s
  8. 08 TinyLlama-1.1B (1T token checkpoint) 1.1B · Q8_0 · 1.9 GB 689 tok/s
  9. 09 TinyLlama-1.1B (3T token checkpoint) 1.1B · Q8_0 · 1.9 GB 689 tok/s
  10. 10 DeciCoder-1B 1.1B · Q8_0 · 1.9 GB 689 tok/s

Step by step

How to work out the tokens per second of a RTX PRO 6000 Blackwell Server

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

    Start with the model, not the specification

    Every one of the 609 models this RTX PRO 6000 Blackwell Server runs is in the table above. Search narrows it by name or by size.

  2. 02

    Set the context length you will actually use

    Drag the slider to the conversation length you plan to work at. The cache grows with the conversation, and on 96 GB it is often what pushes a large model over the edge.

  3. 03

    Set a minimum quality if you need one

    Each model is shown at the best compression this card can hold. A minimum quality hides the ones that only fit by being squeezed further than you would accept.

  4. 04

    Read the speed and the range

    Speeds come with error bars for a reason. The best case here is 758 tok/s on Gemma 3 QAT 1B, and which inference software you use moves that by thirty to fifty per cent.

  5. 05

    Check the headroom before you decide

    Compare what each model needs with the 96 GB this card provides. Tight means it works today; comfortable means it still works when the conversation grows.

  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, and how the RTX PRO 6000 Blackwell Server compares.

Answers

RTX PRO 6000 Blackwell Server — common questions

01

How much power does a RTX PRO 6000 Blackwell Server use?

The RTX PRO 6000 Blackwell Server has a rated board power of 600 W, and a 1,000 W system power supply is suggested. Generating text draws hard in bursts and idles between requests, so average consumption over a working session is normally well below the rated figure.

02

How much cache does a RTX PRO 6000 Blackwell Server have?

The RTX PRO 6000 Blackwell Server has 128 KB of L1 cache, and 128 MB of L2 cache. 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.

03

What are the TFLOPS of a RTX PRO 6000 Blackwell Server?

The RTX PRO 6000 Blackwell Server is rated at 126 TFLOPS at half precision and 126 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.

04

How many tensor cores does a RTX PRO 6000 Blackwell Server have?

The RTX PRO 6000 Blackwell Server has 752 tensor cores across 188 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.

05

Does the RTX PRO 6000 Blackwell Server support CUDA?

Yes. The RTX PRO 6000 Blackwell Server reports CUDA compute capability 12.0. 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.

06

What bus interface does the RTX PRO 6000 Blackwell Server use?

It uses PCIe 5.0 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.

07

Is the RTX PRO 6000 Blackwell Server good for running local AI models?

Its memory is large enough for models most desktop hardware cannot touch and its bandwidth is high enough to generate text faster than most people read. In total it runs 609 of the models we track. Whether that is enough depends entirely on which model you want — the table above answers that directly.

08

Can a RTX PRO 6000 Blackwell Server run a model that does not fit in its memory?

Only partly. Layers beyond the 96 GB sit in system memory and run at a fraction of the speed, so a mostly-offloaded model is rarely worth using. Every figure here assumes it is fully resident on the card.

09

Would two RTX PRO 6000 Blackwell Server cards be twice as fast?

Pairing RTX PRO 6000 Blackwell Server cards buys headroom rather than pace: 192 GB of combined memory, at roughly the same generation speed as one.

10

What AI models can a RTX PRO 6000 Blackwell Server run?

609 of the 679 open-weight language models we track fit on a RTX PRO 6000 Blackwell Server and can be run locally on it. The table on this page lists every one, with the memory it needs, the quantisation it runs at and an estimated generation speed.

11

What is the largest AI model a RTX PRO 6000 Blackwell Server can run?

The largest model in our catalogue that fits on a RTX PRO 6000 Blackwell Server is dots.llm1 at 142B parameters, compressed to IQ4_XS. It generates roughly 13.1 tokens per second and needs about 78.3 GB of the card's memory.

12

How many tokens per second does a RTX PRO 6000 Blackwell Server produce?

It depends on the model. On a RTX PRO 6000 Blackwell Server the fastest model we track is Gemma 3 QAT 1B at about 758 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.

13

Can a RTX PRO 6000 Blackwell Server run a 7B model?

Yes. For example a RTX PRO 6000 Blackwell Server runs Multi-Token Prediction 7B at Q8_0, using about 7.9 GB of memory and generating around 113 tokens per second.

14

Can a RTX PRO 6000 Blackwell Server run a 13B model?

Yes. For example a RTX PRO 6000 Blackwell Server runs DeepSeekMoE-16B at Q8_0, using about 17.5 GB of memory and generating around 263 tokens per second.

15

Can a RTX PRO 6000 Blackwell Server run a 30B model?

Yes. For example a RTX PRO 6000 Blackwell Server runs ERNIE-4.5-VL-28B-A3B at Q8_0, using about 29.2 GB of memory and generating around 150 tokens per second.

16

Can a RTX PRO 6000 Blackwell Server run a 70B model?

Yes. For example a RTX PRO 6000 Blackwell Server runs Qwen3-Coder-Next at Q8_0, using about 80.7 GB of memory and generating around 52.7 tokens per second.

17

How much memory does a RTX PRO 6000 Blackwell Server have?

A RTX PRO 6000 Blackwell Server has 96 GB of GDDR7 memory. Around a tenth of that is reserved by the inference runtime and the driver, leaving roughly 86.4 GB available for a model and its conversation.

18

What is the memory bandwidth of a RTX PRO 6000 Blackwell Server?

The RTX PRO 6000 Blackwell Server has 1,790 GB/s of memory bandwidth, across a 512-bit memory bus. 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.

19

What type of memory does a RTX PRO 6000 Blackwell Server use?

It uses GDDR7 clocked at 1.75 GHz. 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.

20

Who makes the RTX PRO 6000 Blackwell Server?

The RTX PRO 6000 Blackwell Server is a NVIDIA product, with the chip manufactured by TSMC, on a 5 nm process.

21

When was the RTX PRO 6000 Blackwell Server released?

The RTX PRO 6000 Blackwell Server was released in March 2025.

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