can-i-run-this-llm

Can I run this LLM?

Can a AMD Instinct MI300X (192GB) run Qwen3 30B-A3B (MoE)?

Qwen · Apache 2.0 · estimated for a 192.0 GB machine at 4K context.

Yes — it runs great.

Best fit: Qwen3 30B-A3B (MoE) at FP16 · 61.1 GB · 229–344 tok/s on a AMD Instinct MI300X (192GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K9.9 GB325–487 tok/sRuns great
Q3_K_M13.2 GB317–475 tok/sRuns great
Q4_K_M18.5 GB304–456 tok/sRuns great
Q5_K_M21.8 GB296–445 tok/sRuns great
Q6_K25.2 GB289–433 tok/sRuns great
Q8_032.4 GB275–412 tok/sRuns great
FP1661.1 GB229–344 tok/sRuns great

Ran Qwen3 30B-A3B (MoE) on a AMD Instinct MI300X (192GB)? Add your real numbers →