can-i-run-this-llm

Can I run this LLM?

Can a NVIDIA RTX 4090 (24GB) run Qwen3 Coder 30B-A3B (MoE)?

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

Yes — it runs great.

Best fit: Qwen3 Coder 30B-A3B (MoE) at Q3_K_M · 14.7 GB · 192–288 tok/s on a NVIDIA RTX 4090 (24GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K11.3 GB210–316 tok/sRuns great
Q3_K_M14.7 GB192–288 tok/sRuns great
Q4_K_M18.6 GB175–262 tok/sTight / slow
Q5_K_M21.7 GB163–245 tok/sWon't fit
Q6_K25.1 GB152–228 tok/sWon't fit
Q8_032.5 GB132–199 tok/sWon't fit
FP1661.1 GB88–132 tok/sWon't fit
Open the full calculator →Everything a NVIDIA RTX 4090 (24GB) runsQwen3 Coder 30B-A3B (MoE) on other machinesWhat to buy to run it →

Ran Qwen3 Coder 30B-A3B (MoE) on a NVIDIA RTX 4090 (24GB)? Add your real numbers →