Can Phi 3 Medium 14B run on Quadro RTX 8000 48GB?

YES — Runs Great

B59Good

○Estimated from fit model

Phi 3 Medium 14B needs ~17.6 GB VRAM. Quadro RTX 8000 48GB has 48.0 GB. With Q4_K_M quantization, expect ~58 tok/s.

Runtime: OllamaCapacity: RoomyBandwidth: MediumStack: 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.

Capabilities:

Select quantization to explore

Q4_K_M (Medium quality) — 17.6 GB, 58.4 tok/s, Runs well

17.6 GB required48.0 GB available

37% VRAM used

Fit status

Runs well

Decode

58.4 tok/s

TTFT

3317 ms

Safe context

128K

Memory

17.6 GB / 48.0 GB

Memory breakdown

Weights8.5 GB

KV Cache3.1 GB

Runtime1.2 GB

Headroom4.8 GB

See how fast it feels

See how fast it feelsPhi 3 Medium 14B on Quadro RTX 8000 48GB
1st promptCold start — includes initialization
>Solve: If a train travels at 60 mph for 2.5 hours, how far does it go?
Let me work through this step by step.

Given information:
 - Speed = 60 mph (miles per hour)
 - Time = 2.5 hours

Formula: Distance = Speed x Time

Calculation:
 Distance = 60 mph x 2.5 h
 = 150 miles

The train travels 150 miles.

Double-check: at 60 mph the train covers 60 miles each hour. In half an hour it covers 30 miles. So in 2.5 hours: 60 + 60 + 30 = 150 miles. Confirmed.
2nd promptWarm — model ready, no init overhead
>What if the train stops for 20 minutes halfway?
The stop doesn't change the total distance covered while moving, but it affects total trip time.

Breakdown:
 - First half: 1.25 hours at 60 mph = 75 miles
 - Stop: 20 minutes (no distance)
 - Second half: 1.25 hours at 60 mph = 75 miles

Total distance: 150 miles (same)
Total time: 2.5 h + 0.33 h = 2.83 hours
Average speed: 150 / 2.83 = 53 mph

The distance stays the same but average speed drops to 53 mph because of the stop.
Estimated: 58.4 tok/s decode · 3.3s TTFT (warm) · 146 tok/s prefill

What limits this setup

This setup is broadly balanced for this model.

Older PCIe generation

PCIe 3.0 is workable, but it compounds the penalty when you offload heavily or try to scale across multiple cards.

Best improvement path

Performance by workload

Workload	Grade	Fit	Decode	TTFT	Context
Chat	B	Runs well	58.4 tok/s	1809 ms	128K
Coding	B	Runs well	58.4 tok/s	3317 ms	128K
Agentic Coding	B	Runs well	58.4 tok/s	4825 ms	128K
Reasoning	B	Runs well	54.3 tok/s	4214 ms	128K
RAG	B	Runs well	58.4 tok/s	6031 ms	128K

Quantization options

How Phi 3 Medium 14B (14B params) fits at each quantization level on Quadro RTX 8000 48GB (48.0 GB usable).

Quant	Bits	VRAM	Quality	Fit
Q2_K	2	5.5 GB	Low	C53
Q3_K_S	3	6.9 GB	Low	C53
NVFP4	4

Get started

Copy-paste commands to run Phi 3 Medium 14B on your machine.

Run

ollama run phi3:medium

Upgrade options

Hardware that runs Phi 3 Medium 14B well

AMD Instinct MI210 64GBBudget pick

64 GB VRAM (+16)1638 GB/s (+966)

Raises estimated decode speed by about 140%.140.2 tok/s decode

Raises estimated decode speed by about 140%.

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

~$10,000 MSRP

Frequently asked questions

See all results for Quadro RTX 8000 48GB See all hardware for Phi 3 Medium 14B

URL: https://willitrunai.com/can-run/phi-3-medium-14b-on-quadro-rtx-8000-48gb

⇱