How much VRAM does Nemotron Mini 4B need?

Nemotron Mini 4B (4B parameters) requires approximately 5.0 GB of memory with Q4_K_M quantization.

What is the best quantization for Nemotron Mini 4B?

The recommended quantization for Nemotron Mini 4B is Q4_K_M, which balances quality and memory efficiency.

Can it run?

Yes, GTX 1660 Ti 6GB can run Nemotron Mini 4B with a C grade (Tight fit). Expected decode speed: 64.9 tok/s.

CUsable

Tight fit

Using Q4_K_M in Ollama

Capabilities:

Fit status

Tight fit

Decode

64.9 tok/s

TTFT

2982 ms

Safe context

Memory

5.0 GB / 6.0 GB

Weights2.4 GB

KV Cache0.8 GB

Runtime1.2 GB

Headroom0.6 GB

Workload	Grade	Fit	Decode	TTFT	Context
Agentic Coding	C	Tight fit	64.9 tok/s	4338 ms	4K
Chat	C	Tight fit	85.0 tok/s	1243 ms	4K
Coding	C	Tight fit	64.9 tok/s	2982 ms	4K
RAG	C	Tight fit	64.9 tok/s	5422 ms	4K
Reasoning	C	Tight fit	64.9 tok/s	3524 ms	4K

How Nemotron Mini 4B (4B params) fits at each quantization level on GTX 1660 Ti 6GB (6.0 GB usable).

Quant	Bits	VRAM	Quality	Fit
Q2_K	2	1.6 GB	Low	D34
Q3_K_S	3	2.0 GB	Low	D35
NVFP4	4

Ollama

ollama run nemotron-mini-4b

HuggingFace

huggingface-cli download nemotron-mini-4b

Upgrade options

RX 7600 8GBBudget pick

C68.5 tok/s decode

~$269 MSRP

Intel Arc A750 8GBBest value

C90.2 tok/s decode

~$289 MSRP

RTX 2060 Super 8GBNVIDIA upgrade

C106.5 tok/s decode

~$399 MSRP

Intel Arc A580 8GBBiggest leap