can-i-run-this-llm

Can I run this LLM?

Can a AMD Instinct MI300X (192GB) run Qwen3.8 Flash Next (MoE)?

Qwen · See model card · estimated for a 192.0 GB machine at 4K context.

Yes — it runs great.

Best fit: Qwen3.8 Flash Next (MoE) at Q5_K_M · 134.7 GB · 255–383 tok/s on a AMD Instinct MI300X (192GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K80.9 GB289–434 tok/sRuns great
Q3_K_M92.0 GB282–422 tok/sRuns great
Q4_K_M119.6 GB264–396 tok/sRuns great
Q5_K_M134.7 GB255–383 tok/sRuns great
Q6_K168.1 GB238–356 tok/sTight / slow
Q8_0188.3 GB228–342 tok/sWon't fit
FP16354.0 GB172–258 tok/sWon't fit
Open the full calculator →Everything a AMD Instinct MI300X (192GB) runsQwen3.8 Flash Next (MoE) on other machinesWhat to buy to run it →

Ran Qwen3.8 Flash Next (MoE) on a AMD Instinct MI300X (192GB)? Add your real numbers →