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

Qwen3-Coder 30B A3B Instruct (30.5B parameters) requires approximately 22.6 GB of memory with Q4_K_M quantization.

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

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

Can it run?

Can RX 7900 XT 20GB run Qwen3-Coder 30B A3B Instruct?

Q: Can RX 7900 XT 20GB run Qwen3-Coder 30B A3B Instruct?

Yes, RX 7900 XT 20GB can run Qwen3-Coder 30B A3B Instruct with a C grade (Very compromised (needs ~2.1 GB host RAM)). Expected decode speed: 60.2 tok/s.

CUsable

Very compromised (needs ~2.1 GB host RAM)

Using Q4_K_M in Ollama

Capabilities:

Fit status

Very compromised (needs ~2.1 GB host RAM)

Decode

60.2 tok/s

TTFT

3216 ms

Safe context

14K

Memory

22.6 GB / 20.0 GB

Offload

10%

Memory breakdown

Weights18.6 GB

KV Cache0.8 GB

Runtime1.2 GB

Headroom2.0 GB

Performance by workload

Workload	Grade	Fit	Decode	TTFT	Context
Agentic Coding	C	Very compromised (needs ~2.3 GB host RAM)	59.7 tok/s	4717 ms	28K
Chat	C	Very compromised (needs ~2.1 GB host RAM)	60.2 tok/s	1754 ms	7K
Coding	C	Very compromised (needs ~2.1 GB host RAM)	60.2 tok/s	3216 ms	14K
RAG	C	Very compromised (needs ~2.3 GB host RAM)	59.7 tok/s	5897 ms	28K
Reasoning	C	Very compromised (needs ~2.1 GB host RAM)	60.2 tok/s

Quantization options

How Qwen3-Coder 30B A3B Instruct (30.5B params) fits at each quantization level on RX 7900 XT 20GB (20.0 GB usable).

Quant	Bits	VRAM	Quality	Fit
Q2_KBest for your GPU	2	11.9 GB	Low	C42
Q3_K_S	3	14.9 GB	Low	C45

Get started

Ollama

ollama run qwen-3-coder-30b-a3b

HuggingFace

huggingface-cli download qwen-3-coder-30b-a3b

Upgrade options

Hardware that runs Qwen3-Coder 30B A3B Instruct well

MacBook Pro M4 32GBBest value

C11.9 tok/s decode

~$799 MSRP

Mac mini M4 64GBBudget pick

C12 tok/s decode

~$1,099 MSRP

See all results for RX 7900 XT 20GB See all hardware for Qwen3-Coder 30B A3B Instruct