Can EXAONE 4.0 32B run on Intel Arc A770 16GB?

YES — With Q2_K

B66Good

Estimated from fit model

EXAONE 4.0 32B needs ~18.9 GB VRAM. Intel Arc A770 16GB has 16.0 GB. With Q2_K quantization, expect ~10 tok/s.

Runtime: llama.cppCapacity: OffloadBandwidth: MediumStack: StandardBottleneck: Host offload

Operating mode

Choose the run profile you care about

Interactive favors responsiveness, while light API and scale-out lean harder on serving readiness. The fit stays the same, but the recommendation lens changes.

Current mode

Balanced

Balanced for general local use. Keeps the ranking neutral across personal and serving workflows.

EXAONE 4.0 32B at Q4_K_M needs 25.9 GB — too much for Intel Arc A770 16GB (16.0 GB). Runs at Q2_K (18.9 GB) with low quality.

Capabilities:

Select quantization to explore

Q4_K_M (Medium quality) — 25.9 GB, exceeds 16.0 GB available

25.9 GB required16.0 GB available

162% VRAM needed

9.9 GB over capacity — needs offload or smaller quantization

Fit status

Too heavy

Decode

3.8 tok/s

TTFT

51139 ms

Safe context

Memory

25.9 GB / 16.0 GB

Offload

40%

Memory breakdown

Weights19.5 GB

KV Cache3.9 GB

Runtime0.9 GB

Headroom1.6 GB

See how fast it feels

With memory offload — actual speed may be lower

See how fast it feelsEXAONE 4.0 32B on Intel Arc A770 16GB
1st promptCold start — includes initialization
>Write a Python function to sort a list
Here's a Python function that sorts a list using quicksort:

def quicksort(arr):
    if len(arr) <= 1:
        return arr
    pivot = arr[len(arr) // 2]
    left   = [x for x in arr if x < pivot]
    middle = [x for x in arr if x == pivot]
    right  = [x for x in arr if x > pivot]
    return quicksort(left) + middle + quicksort(right)

# Example usage
numbers = [3, 6, 8, 10, 1, 2, 1]
print(quicksort(numbers))  # [1, 1, 2, 3, 6, 8, 10]

The algorithm picks a pivot, partitions the list into smaller, equal, and larger elements, then recursively sorts each partition and concatenates the results.
2nd promptWarm — model ready, no init overhead
>Now add type hints and a docstring
def quicksort(arr: list[int]) -> list[int]:
    """Sort a list using the quicksort algorithm.

    Args:
        arr: List of integers to sort.

    Returns:
        A new sorted list.
    """
    if len(arr) <= 1:
        return arr
    pivot = arr[len(arr) // 2]
    left   = [x for x in arr if x < pivot]
    middle = [x for x in arr if x == pivot]
    right  = [x for x in arr if x > pivot]
    return quicksort(left) + middle + quicksort(right)

Added type hints for the input and return type, plus a Google-style docstring.
Estimated: 3.8 tok/s decode · 51.1s TTFT (warm) · 10 tok/s prefill

What limits this setup

It fits through host-memory offload, and offload is the main reason performance drops.

CPU or host-memory offload is active

About 20% of the working set spills out of accelerator memory, which usually hurts latency and sustained decode throughput.

Very little memory headroom

You can run the model, but there is not much room left for longer context, bigger batches, extra apps, or future model updates.

Runtime ecosystem is narrower than CUDA

Intel GPUs can look attractive on memory per dollar, but local AI tooling, kernels, and model coverage are still broader and easier on CUDA today.

Best improvement path

Remove offload with more accelerator memory

Prioritize a GPU or unified-memory tier that fits the whole model natively. Removing offload usually helps more than small compute gains.

Prefer CUDA if you want the path of least resistance

If your goal is maximum runtime coverage, easier troubleshooting, and better support for new local AI releases, CUDA is usually still the safer upgrade path.

Buy headroom, not only minimum fit

A slightly larger memory tier gives you safer context growth and makes the recommendation more future-proof.

Performance by workload

Workload	Grade	Fit	Decode	TTFT	Context
Chat	F	Too heavy	4.5 tok/s	23654 ms	4K
Coding	F	Too heavy	3.8 tok/s	51139 ms	4K
Agentic Coding	F	Too heavy	2.8 tok/s	99948 ms	4K
Reasoning	F	Too heavy	3.8 tok/s	60437 ms	4K
RAG	F	Too heavy	2.8 tok/s	124936 ms	4K

Inference speed

EXAONE 4.0 32B inference speed — tokens per second by GPU & Mac

Estimated decode speed (tokens/sec) for EXAONE 4.0 32B at Q4_K_M across popular GPUs and Apple Silicon, using the fastest local runtime per device. Fastest is RTX 5090 32GB at ~66 tok/s. Speed is memory-bandwidth bound, so cards that fit the whole model in VRAM run far faster than ones that offload to system RAM.

GPU / Mac	Memory	Quant	Speed (tok/s)	Fits?
RTX 5090 32GB	32 GB	Q4_K_M	66.4	Tight
MacBook Pro M4 Max 128GB	128 GB	Q4_K_M	33.2	Fits
MacBook Pro M4 Max 64GB	64 GB	Q4_K_M	33.2	Fits
Mac Studio M3 Ultra 256GB	256 GB	Q4_K_M	30.8	Fits
Mac Studio M2 Ultra 128GB	128 GB	Q4_K_M	25.7	Fits
RTX 4090 24GB	24 GB	Q4_K_M	24.8	Heavy offload
Mac Studio M1 Ultra 128GB	128 GB	Q4_K_M	24.3	Fits
RX 7900 XTX 24GB	24 GB	Q4_K_M	22.9	Heavy offload
RTX 3090 24GB	24 GB	Q4_K_M	21.2	Heavy offload
MacBook Pro M4 Pro 48GB	48 GB	Q4_K_M	20.9	Tight
MacBook Pro M3 Max 64GB	64 GB	Q4_K_M	13.3	Fits
MacBook Pro M1 Max 64GB	64 GB	Q4_K_M	12.2	Fits
RTX 4080 Super 16GB	16 GB	Q4_K_M	9.0	Too big
RTX 4070 12GB	12 GB	Q4_K_M	3.1	Too big
RTX 3060 12GB	12 GB	Q4_K_M	2.0	Too big
RTX 4060 8GB	8 GB	Q4_K_M	2.0	Too big

Estimates for single-stream decoding at Q4_K_M; real tokens/sec varies with prompt length, context, batch size, and runtime build. Prompt processing (prefill) is faster than the decode figures shown here.

Quantization options

How EXAONE 4.0 32B (32B params) fits at each quantization level on Intel Arc A770 16GB (16.0 GB usable).

Quant	Bits	VRAM	Quality	Fit
Q2_K	2	12.5 GB	Low	F0
Q3_K_S	3	15.7 GB	Low	F0
NVFP4	4	17.9 GB	Medium	F0
Q4_K_M	4	19.5 GB	Medium	F0
Q5_K_M	5	23.0 GB	High	F0
Q6_K	6	26.2 GB	High	F0
Q8_0	8	34.2 GB	Very High	F0
F16	16	65.6 GB	Maximum	F0

Get started

Copy-paste commands to run EXAONE 4.0 32B on your machine.

Run

ollama run exaone-4:32b

アップグレードオプション

EXAONE 4.0 32Bを快適に動かすハードウェア

Intel Arc Pro B60 24GBコスパ最良

24 GB VRAM (+8)

Makes the model fit on the accelerator instead of staying completely out of reach.8.4 tok/sデコード

Makes the model fit on the accelerator instead of staying completely out of reach.

Raises estimated decode speed by about 121%.

〜$599 MSRP

NVIDIA A100 40GB最大の飛躍

40 GB VRAM (+24)1555 GB/s (+995)

Makes the model fit on the accelerator instead of staying completely out of reach.72.3 tok/sデコード

Makes the model fit on the accelerator instead of staying completely out of reach.

Removes host-memory offload, which is usually the single biggest latency and throughput win.

〜$10,000 MSRP

Intel Data Center GPU Max 1550 128GBコスパ重視

128 GB VRAM (+112)3200 GB/s (+2640)

Makes the model fit on the accelerator instead of staying completely out of reach.111.5 tok/sデコード

Makes the model fit on the accelerator instead of staying completely out of reach.

Removes host-memory offload, which is usually the single biggest latency and throughput win.

〜$15,000 MSRP

Gaudi 3 128GBIntelアップグレード

128 GB VRAM (+112)3700 GB/s (+3140)

Makes the model fit on the accelerator instead of staying completely out of reach.143.3 tok/sデコード

Makes the model fit on the accelerator instead of staying completely out of reach.

Removes host-memory offload, which is usually the single biggest latency and throughput win.

〜$15,000 MSRP

Frequently asked questions

Can Intel Arc A770 16GB run EXAONE 4.0 32B?

Yes, Intel Arc A770 16GB can run EXAONE 4.0 32B at Q2_K quantization (Very compromised (needs ~1.9 GB host RAM)). The recommended Q4_K_M requires 25.9 GB which exceeds available memory, but at Q2_K it needs only 18.9 GB. Expected decode speed: 9.8 tok/s.

How much VRAM does EXAONE 4.0 32B need?

EXAONE 4.0 32B (32B parameters) requires approximately 25.9 GB at Q4_K_M quantization. On Intel Arc A770 16GB, it fits at Q2_K using 18.9 GB.

What is the best quantization for EXAONE 4.0 32B?

The recommended quantization is Q4_K_M, but on Intel Arc A770 16GB the best fitting quantization is Q2_K, which uses 18.9 GB.

What speed will EXAONE 4.0 32B run at on Intel Arc A770 16GB?

On Intel Arc A770 16GB, EXAONE 4.0 32B achieves approximately 9.8 tokens per second decode speed with a time-to-first-token of 19744ms using Q2_K quantization.

Can Intel Arc A770 16GB run EXAONE 4.0 32B for coding?

For coding workloads, EXAONE 4.0 32B on Intel Arc A770 16GB receives a F grade with 3.8 tok/s and 4K context.

What context window can EXAONE 4.0 32B use on Intel Arc A770 16GB?

On Intel Arc A770 16GB, EXAONE 4.0 32B can safely use up to 4K tokens of context at Q2_K quantization. The model's official context limit is 131K, but available memory constrains the safe maximum.

What should I upgrade first if EXAONE 4.0 32B feels slow on Intel Arc A770 16GB?

Remove offload with more accelerator memory. Prioritize a GPU or unified-memory tier that fits the whole model natively. Removing offload usually helps more than small compute gains.

Would CUDA be a better path than Intel Arc A770 16GB for EXAONE 4.0 32B?

Often yes, if your goal is the easiest setup and the widest runtime support. Intel can offer attractive memory capacity, but CUDA still tends to win on tooling maturity, guides, kernels, and model coverage for local AI.

See all results for Intel Arc A770 16GB See all hardware for EXAONE 4.0 32B

Embed this result▼

Paste this snippet into any page to show a live fit card.

<iframe src="https://willitrunai.com/embed/exaone-4-32b-on-arc-a770-16gb" width="400" height="180" frameborder="0" style="border:none;border-radius:12px;overflow:hidden;" title="Will It Run AI — fit result"></iframe>

Preview: