How much VRAM does Hermes 4.3 36B need?

Hermes 4.3 36B (36B parameters) requires approximately 35.7 GB of memory with Q4_K_M quantization.

What is the best quantization for Hermes 4.3 36B?

The recommended quantization for Hermes 4.3 36B is Q4_K_M, which balances quality and memory efficiency.

Can it run?

Yes, Mac mini M4 64GB can run Hermes 4.3 36B with a C grade (Runs well). Expected decode speed: 3.9 tok/s.

CUsable

Runs well

Using Q4_K_M in Ollama

Capabilities:

Fit status

Runs well

Decode

3.9 tok/s

TTFT

49207 ms

Safe context

21K

Memory

35.7 GB / 46.1 GB

Weights22.0 GB

KV Cache5.6 GB

Runtime1.2 GB

Headroom6.9 GB

Workload	Grade	Fit	Decode	TTFT	Context
Agentic Coding	C	Tight fit	3.9 tok/s	71573 ms	36K
Chat	C	Runs well	3.9 tok/s	26840 ms	11K
Coding	C	Runs well	3.9 tok/s	49207 ms	21K
RAG	C	Tight fit	3.9 tok/s	89467 ms	36K
Reasoning	C	Runs well	3.9 tok/s	58153 ms	21K

How Hermes 4.3 36B (36B params) fits at each quantization level on Mac mini M4 64GB (46.1 GB usable).

Quant	Bits	VRAM	Quality	Fit
Q2_K	2	14.0 GB	Low	D35
Q3_K_S	3	17.6 GB	Low	D37
NVFP4	4

Upgrade options

MacBook Pro M4 Max 96GBBudget pick

C15.7 tok/s decode

~$2,499 MSRP

MacBook Pro M2 Max 96GBBest value

C10.6 tok/s decode

~$3,199 MSRP

RTX PRO 5000 Blackwell 48GBBiggest leap