Question 1

Can NVIDIA B200 180GB run Llama 3.3 70B?

Accepted Answer

Yes, NVIDIA B200 180GB can run Llama 3.3 70B with a C grade (Runs well). Expected decode speed: 157.4 tok/s.

Question 2

How much VRAM does Llama 3.3 70B need?

Accepted Answer

Llama 3.3 70B (70B parameters) requires approximately 72.8 GB of memory with Q4_K_M quantization.

Question 3

What is the best quantization for Llama 3.3 70B?

Accepted Answer

The recommended quantization for Llama 3.3 70B is Q4_K_M, which balances quality and memory efficiency.

Workload	Grade	Fit	Decode	TTFT	Context
Agentic Coding	C	Runs well	157.4 tok/s	1789 ms	69K
Chat	C	Runs well	157.4 tok/s	671 ms	21K
Coding	C	Runs well	157.4 tok/s	1230 ms	40K
RAG	C	Runs well	157.4 tok/s	2237 ms	69K
Reasoning	C	Runs well	157.4 tok/s	1454 ms	40K

Quant	Bits	VRAM	Quality	Fit
Q2_K	2	27.3 GB	Low	D33
Q3_K_S	3	34.3 GB	Low	D34
NVFP4	4

Can NVIDIA B200 180GB run Llama 3.3 70B?

Memory breakdown