VOOZH about

URL: https://willitrunai.com/can-run/hf-lmstudio-community--exaone-3-5-2-4b-instruct-gguf-on-dgx-spark-128gb

⇱ EXAONE 3.5 2.4B Instruct on NVIDIA DGX Spark 128GB? YES

Can EXAONE 3.5 2.4B Instruct run on NVIDIA DGX Spark 128GB?

YES — Runs Great

C42Usable
○Estimated from fit model

EXAONE 3.5 2.4B Instruct needs ~16.0 GB VRAM. NVIDIA DGX Spark 128GB has 108.8 GB. With Q4_K_M quantization, expect ~34 tok/s.

Runtime: OllamaCapacity: RoomyBandwidth: LowStack: BasicBottleneck: Balanced
Share:

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.

Capabilities:

Select quantization to explore

Q4_K_M (Medium quality) — 16.0 GB, 33.6 tok/s, Runs well
16.0 GB required108.8 GB available
15% VRAM used

Fit status

Runs well

Decode

33.6 tok/s

TTFT

5762 ms

Safe context

5.3M

Memory

16.0 GB / 108.8 GB

Memory breakdown

Weights1.5 GB
KV Cache0.3 GB
Runtime1.2 GB
Headroom13.1 GB

See how fast it feels

See how fast it feelsEXAONE 3.5 2.4B Instruct on NVIDIA DGX Spark 128GB
1st promptCold start — includes initialization
>What is local AI inference?
Local AI inference means running an AI language model directly on your own hardware — your laptop, desktop, or server — instead of sending requests to a remote cloud API. When you run inference locally the model weights are loaded into your GPU or unified memory. Each token you generate requires reading those weights from memory, so memory bandwidth is the main bottleneck for decode speed. Key benefits of running locally: - Full privacy: your prompts never leave your machine - No per-token cost or rate limits - Works offline once the model is downloaded - Latency depends only on your hardware
2nd promptWarm — model ready, no init overhead
>How much VRAM do I need?
It depends on the model size and quantization level. A rough rule of thumb: Model size Q4 (4-bit) Q8 (8-bit) FP16 7B params ~4.3 GB ~7.5 GB ~14 GB 13B params ~7.9 GB ~13.9 GB ~26 GB 70B params ~42.7 GB ~74.9 GB ~140 GB Most people use 4-bit quantization (Q4_K_M) which gives 90-95% of full quality at a fraction of the memory. A 24 GB GPU can comfortably run most 7B-13B models.
Estimated: 33.6 tok/s decode · 5.8s TTFT (warm) · 84 tok/s prefill

What limits this setup

This setup is broadly balanced for this model.

Shared-memory contention still exists

The OS, browser, and inference runtime all compete for the same physical memory pool, so real-world headroom is less forgiving than raw capacity suggests.

Best improvement path

Performance by workload

WorkloadGradeFitDecodeTTFTContext
ChatCRuns well33.6 tok/s3143 ms5.3M
CodingCRuns well33.6 tok/s5762 ms5.3M
Agentic CodingCRuns well33.6 tok/s8381 ms5.3M
ReasoningCRuns well33.6 tok/s6810 ms5.3M
RAGCRuns well33.6 tok/s10476 ms5.3M

Quantization options

How EXAONE 3.5 2.4B Instruct (2.4000000953674316B params) fits at each quantization level on NVIDIA DGX Spark 128GB (92.2 GB usable).

QuantBitsVRAMQualityFit
Q2_K
2
0.9 GB
LowD39
Q3_K_S
3
1.2 GB
LowD39
NVFP4
4
1.3 GB
MediumD39
Q4_K_M
4
1.5 GB
MediumD39
Q5_K_M
5
1.7 GB
HighD39
Q6_K
6
2.0 GB
HighD39
Q8_0
8
2.6 GB
Very HighD39
F16Best for your GPU
16
4.9 GB
MaximumD39

Get started

Copy-paste commands to run EXAONE 3.5 2.4B Instruct on your machine.

Run

lms load hf-lmstudio-community--exaone-3-5-2-4b-instruct-gguf && lms server start

Upgrade options

Hardware that runs EXAONE 3.5 2.4B Instruct well

Mac Studio M3 Ultra 256GBBudget pick
256 GB Unified (+128)819 GB/s (+546)
C
Adds memory headroom for longer context windows and future model growth.33.6 tok/s decode

Adds memory headroom for longer context windows and future model growth.

~$6,999 MSRP

Frequently asked questions

See all results for NVIDIA DGX Spark 128GBSee all hardware for EXAONE 3.5 2.4B Instruct