Question 1

Can NVIDIA GH200 96GB run Qwen3-VL 30B A3B Instruct?

Accepted Answer

Yes, NVIDIA GH200 96GB can run Qwen3-VL 30B A3B Instruct with a C grade (Runs well). Expected decode speed: 465.9 tok/s.

Question 2

How much VRAM does Qwen3-VL 30B A3B Instruct need?

Accepted Answer

Qwen3-VL 30B A3B Instruct (30B parameters) requires approximately 31.1 GB of memory with Q4_K_M quantization.

Question 3

What is the best quantization for Qwen3-VL 30B A3B Instruct?

Accepted Answer

The recommended quantization for Qwen3-VL 30B A3B Instruct is Q4_K_M, which balances quality and memory efficiency.

Workload	Grade	Fit	Decode	TTFT	Context
Agentic Coding	C	Runs well	414.2 tok/s	680 ms	98K
Chat	C	Runs well	465.9 tok/s	350 ms	25K
Coding	C	Runs well	465.9 tok/s	416 ms	49K
RAG	C	Runs well	465.9 tok/s	755 ms	98K
Reasoning	C	Runs well	465.9 tok/s	491 ms	49K

Quant	Bits	VRAM	Quality	Fit
Q2_K	2	11.7 GB	Low	D32
Q3_K_S	3	14.7 GB	Low	D33
NVFP4	4

Can NVIDIA GH200 96GB run Qwen3-VL 30B A3B Instruct?

Memory breakdown