Can 4x RTX 3090 run Qwen3 235B-A22B?
Yes. Measured 18 tok/s generation (llama.cpp, IQ4_XS; source).
Every measured result for Qwen3 235B-A22B on 4x RTX 3090
| Build | Quant / setup | Engine | tok/s | |
|---|---|---|---|---|
| 4x RTX 3090 + EPYC 7443, 256 GB DDR4 8-channel | IQ4_XS | llama.cpp | 18 | source ⚠ unchecked |
| 4x RTX 3090 + 32-core Threadripper, 192 GB DDR4 | UD-Q3_K_XL | llama.cpp | 12-14 | source ⚠ unchecked |
Speeds are decode (generation) tokens/sec as reported by each source; quant, engine and context differ, so small gaps are noise.
All measured speeds for Qwen3 235B-A22B: best home rig for Qwen3 235B-A22B.
Not enough VRAM? Rent before you buy. Run Qwen3 235B-A22B on a bigger GPU on a rented 24–96 GB cloud GPU by the hour and see the real speed before spending €1,000+. Try it on Runpod
Referral link: if you sign up with Google and add $10, Runpod gives you and RigBase bonus credit (terms). It never affects our data.
Referral link: if you sign up with Google and add $10, Runpod gives you and RigBase bonus credit (terms). It never affects our data.