can-i-run-this-llm

Can I run this LLM?

Can a NVIDIA RTX 3090 (24GB) run Qwen3.8 27B?

Qwen · Apache 2.0 · estimated for a 24.0 GB machine at 4K context.

Yes — it runs great.

Best fit: Qwen3.8 27B at Q4_K_M · 17.4 GB · 35–53 tok/s on a NVIDIA RTX 3090 (24GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K10.8 GB53–80 tok/sRuns great
Q3_K_M13.4 GB44–67 tok/sRuns great
Q4_K_M17.4 GB35–53 tok/sRuns great
Q5_K_M20.9 GB30–45 tok/sWon't fit
Q6_K23.9 GB27–40 tok/sWon't fit
Q8_029.1 GB22–33 tok/sWon't fit
FP1654.7 GB12–18 tok/sWon't fit

Ran Qwen3.8 27B on a NVIDIA RTX 3090 (24GB)? Add your real numbers →