Can Phi 3.5 Mini 4B run on NVIDIA DGX Spark 128GB?

YES — With F16

B59Good

○Estimated from fit model

Phi 3.5 Mini 4B needs ~28.3 GB VRAM. NVIDIA DGX Spark 128GB has 0 MB. With F16 quantization, expect ~28 tok/s.

Runtime: OllamaCapacity: RoomyBandwidth: LowStack: BasicBottleneck: Balanced

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.

Phi 3.5 Mini 4B at Q4_K_M needs 9.5 GB — too much for NVIDIA DGX Spark 128GB (0.0 GB). Runs at F16 (28.3 GB) with maximum quality. 8 quantization levels fit.

Capabilities:

Select quantization to explore

Q4_K_M (Medium quality) — 22.6 GB, 56.0 tok/s, Runs well

22.6 GB required108.8 GB available

21% VRAM used

Fit status

Runs well

Decode

56.0 tok/s

TTFT

3457 ms

Safe context

128K

Memory

22.6 GB / 108.8 GB

Memory breakdown

Weights2.4 GB

KV Cache5.9 GB

Runtime1.2 GB

Headroom13.1 GB

See how fast it feels

See how fast it feelsPhi 3.5 Mini 4B 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: 56.0 tok/s decode · 3.5s TTFT (warm) · 140 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

Workload	Grade	Fit	Decode	TTFT	Context
Chat	F	Too heavy	12.1 tok/s	8739 ms	4K
Coding	F	Too heavy	12.1 tok/s	16022 ms	4K
Agentic Coding	F	Too heavy	12.1 tok/s	23304 ms	4K
Reasoning	F	Too heavy	12.1 tok/s	18935 ms	4K
RAG	F	Too heavy	12.1 tok/s	29130 ms	4K

Quantization options

How Phi 3.5 Mini 4B (4B params) fits at each quantization level on NVIDIA DGX Spark 128GB (92.2 GB usable).

Quant	Bits	VRAM	Quality	Fit
Q2_K	2	1.6 GB	Low	C55
Q3_K_S	3	2.0 GB	Low	C55
NVFP4	4

Get started

Copy-paste commands to run Phi 3.5 Mini 4B on your machine.

Run

ollama run phi3.5

Frequently asked questions

See all results for NVIDIA DGX Spark 128GB See all hardware for Phi 3.5 Mini 4B

URL: https://willitrunai.com/can-run/phi-3.5-mini-4b-on-dgx-spark-128gb

⇱