Calculate the TPS of the Arc A730M on local AI models

Intel 12 GB GDDR6 336 GB/s January 2022

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

411 models it can run

721 models in our catalogue altogether

Largest model it holds

ERNIE-4.5-21B-A3B

21B · Q3_K_M · 66.0 tok/s

Fastest model

Gemma 3 QAT 1B

92.5 tok/s · 1B

Which AI models can run on a Arc A730M?

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.

411 models match

Calculating
Quantisation Fit
92.5 tok/s

56–148 · low confidence

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

56–148 · low confidence

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

56–148 · low confidence

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

56–148 · low confidence

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

56–148 · low confidence

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

56–148 · low confidence

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

51–137 · low confidence

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

50–135 · low confidence

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

50–135 · low confidence

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

50–135 · low confidence

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

50–135 · low confidence

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

46–123 · low confidence

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

46–123 · low confidence

LFM2-1.2B ≈ 1.2B Jul 2025 2.0 GB 131k tokens ? Q8_0 Comfortable
77.1 tok/s

46–123 · low confidence

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

46–123 · low confidence

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

46–123 · low confidence

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

45–120 · low confidence

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

44–119 · low confidence

DeepSeekMoE-16B 16B Jan 2024 10.0 GB 4k tokens Q4_K_M Tight
74.2 tok/s

44–119 · low confidence

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

43–114 · low confidence

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

43–114 · low confidence

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

43–114 · low confidence

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

43–114 · low confidence

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

43–114 · low confidence

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

43–114 · low confidence

Kosmos-2.5 ≈ 1.3B Aug 2024 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

Arc A730M 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
12 GB
Memory bandwidth
336 GB/s
Memory type
GDDR6
Memory bus width
192 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
DG2-512
Architecture
Xe-HPG
Generation
Alchemist(Arc 7 Mobile)
Foundry
TSMC
Process size
6 nm
Transistors
21.7 billion
Transistor density
53,400 K/mm²
Die size
406 mm²
Released
1 January 2022

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
1.1 GHz
Boost clock
2.05 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
3,072
Texture mapping units
192
Render output units
96
Ray tracing cores
24
L2 cache
12 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)
25.2 TFLOPS
Single precision (FP32)
12.6 TFLOPS
Pixel rate
197 GPixel/s
Texture rate
394 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)
80 W
Bus interface
PCIe 4.0 x16
Slot width
IGP

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.

DirectX
12.2
OpenGL
4.6
Vulkan
1.4
OpenCL
3.0
Shader model
6.6

Listings

Where to buy a Arc A730M

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

Capacity and bandwidth

Memory

12 GB

Bandwidth

336 GB/s

Largest model

ERNIE-4.5-21B-A3B

Arc A730M carries 12 GB of GDDR6. That reaches comfortably into small and mid-sized models, though the largest stay out of reach without splitting them. Driver overhead leaves roughly 10.8 GB.

Memory bandwidth reaches 336 GB/s across a bus of 192 bits. That is the number governing generation speed — arithmetic per byte read is small enough that the bus, not the cores, is what everything waits on.

That comes from a memory clock of 1.75 GHz. It is why core counts predict generation speed so poorly.

The practical ceiling is ERNIE-4.5-21B-A3B, 21B, compressed to Q3_K_M and generating around 66.0 tokens per second.

The chip and how it was built

Arc A730M is built on the graphics processor DG2-512, using the architecture Xe-HPG from Intel, as part of the generation Alchemist(Arc 7 Mobile).

The chip is manufactured by TSMC, on a process of 6 nm, with a die measuring 406 mm², holding 21.7 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 January 2022, roughly 4.70051867674 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

25.2 TFLOPS

On paper Arc A730M reaches 25.2 TFLOPS at half precision, and 12.6 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.

Clocks run from a base of 1.1 GHz to a boost of 2.05 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

backed by an L2 cache of 12 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 3,072 shading units, 192 texture mapping units, and 96 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

80 W

Arc A730M is rated at 80 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 igp. 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 4.0 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 Arc A730M

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 ERNIE-4.5-21B-A3B 21B · Q3_K_M · Jun 2025 66.0 tok/s
  2. 02 GigaChat Lite (GigaChat-20B-A3B) 20B · Q3_K_M · Dec 2024 69.3 tok/s
  3. 03 InternLM2.5 20B · Q3_K_M · Aug 2024 12.5 tok/s
  4. 04 Granite 20B 20B · Q3_K_M · May 2024 12.5 tok/s
  5. 05 InternLM2-20B 20B · Q3_K_M · Jan 2024 12.5 tok/s
  6. 06 CogAgent 18B · IQ4_XS · Dec 2023 12.6 tok/s
  7. 07 SPHINX (Llama 2 13B) 19.9B · Q3_K_M · Nov 2023 12.5 tok/s
  8. 08 CogVLM-17B 17B · IQ4_XS · Nov 2023 13.4 tok/s
  9. 09 Flan UL2 19.5B · Q3_K_M · Mar 2023 12.8 tok/s
  10. 10 Palmyra Large 20B 20B · Q3_K_M · Mar 2023 12.5 tok/s

The fastest AI models on a Arc A730M

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 92.5 tok/s
  2. 02 Gemma 3 1B 1B · Q8_0 · 1.8 GB 92.5 tok/s
  3. 03 LLama 3..2 Typhoon 2 1B 1B · Q8_0 · 1.8 GB 92.5 tok/s
  4. 04 OLMo-1B 1B · Q8_0 · 1.8 GB 92.5 tok/s
  5. 05 HGRN 1B (WT 103) 1B · Q8_0 · 1.8 GB 92.5 tok/s
  6. 06 Pythia-1b 1B · Q8_0 · 1.8 GB 92.5 tok/s
  7. 07 OpenELM-1.1B 1.1B · Q8_0 · 1.9 GB 85.7 tok/s
  8. 08 TinyLlama-1.1B (1T token checkpoint) 1.1B · Q8_0 · 1.9 GB 84.1 tok/s
  9. 09 TinyLlama-1.1B (3T token checkpoint) 1.1B · Q8_0 · 1.9 GB 84.1 tok/s
  10. 10 DeciCoder-1B 1.1B · Q8_0 · 1.9 GB 84.1 tok/s

Step by step

How to work out the tokens per second of a Arc A730M

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

    Search for the model you want

    The table lists 411 models this card runs. Search narrows the list by name or by size.

  2. 02

    Decide how long your conversations run

    Drag the slider to the conversation length you plan to work at. The cache grows with the conversation, and against a card holding 12 GB that is frequently the difference between a model fitting and not.

  3. 03

    Pin the comparison to one quality level

    Compression is what lets bigger models fit. The quality control drops any model that needs more of it than you are willing to give.

  4. 04

    Read the speed and the range

    Speeds come with error bars for a reason. The best case here is 92.5 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 headroom before you decide

    Tight means it works today; comfortable means it still works when the conversation grows. Compare what each model needs against an available 12 GB.

  6. 06

    Cross-check against other hardware

    Following a model through to its own page lists all the hardware that can run it, so you can see how it compares against Arc A730M.

Answers

Arc A730M — common questions

01

Arc A730M— 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 92.5 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.

02

Arc A730M— can it run 7B models?

Yes. For example it runs Gemma 4 E4B at Q8_0, using about 9.8 GB of memory and generating around 20.6 tokens per second.

03

Arc A730M— can it run 13B models?

Yes. For example it runs DeepSeekMoE-16B at Q4_K_M, using about 10.0 GB of memory and generating around 74.2 tokens per second.

04

Arc A730M— how much memory does it have?

This card has 12 GB of GDDR6. Around a tenth is reserved by the inference runtime and the driver, leaving roughly 10.8 GB available for a model and its conversation.

05

Arc A730M— what is its memory bandwidth?

Memory bandwidth reaches 336 GB/s across a bus of 192 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.

06

Arc A730M— what type of memory does it use?

It uses GDDR6 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.

07

Arc A730M— who makes it?

This is a product of Intel, with the chip manufactured by TSMC, on a process of 6 nm.

08

Arc A730M— when was it released?

It was released in January 2022.

09

Arc A730M— how much power does it use?

Rated board power is 80 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.

10

Arc A730M— how much cache does it have?

and the L2 cache is 12 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.

11

Arc A730M— what are its TFLOPS?

It is rated at 25.2 TFLOPS at half precision and 12.6 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.

12

Arc A730M— does it support CUDA?

No. CUDA is NVIDIA-only, and this is a card from Intel. It runs language models through ROCm, Vulkan or Metal depending on the software, which are less mature than the CUDA path — our estimates apply a penalty for that.

13

Arc A730M— what bus interface does it use?

It uses PCIe 4.0 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.

14

Arc A730M— is it good for running local AI models?

Its memory covers small and mid-sized models, though the largest are out of reach though its bandwidth means generation will feel slow on larger models. In total it runs 411 of the models we track. Whether that is enough depends entirely on which model you want — the table above answers that directly.

15

Arc A730M— can it run a model that does not fit in its memory?

It can be split, with the overflow held in system memory beyond the card's 12 GB drags the whole thing down, and none of the figures on this page assume it.

16

Would two Arc A730M cards be twice as fast?

No. A second card doubles the memory to 24 GB to work with rather than twice the tokens per second — every figure here is for a single card.

17

Arc A730M— which AI models can it run?

411 of the 721 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.

18

Arc A730M— what is the largest AI model it can run?

The largest model in our catalogue that fits is ERNIE-4.5-21B-A3B at 21B parameters, compressed to Q3_K_M. It generates roughly 66.0 tokens per second and needs about 10.1 GB of the card's memory.

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