Qwen on NVIDIA Ada

Can I run Qwen3 30B-A3B on an RTX 4070 Super?

Not at Q4_K_M — it needs 18.8 GB against 10.5 GB available. You would need 2 of these cards.

Won't fit

18.8 GB of 10.5 GB · 180%
020 GB
Weights 17.3 GB
KV cache 0.8 GB
Runtime overhead 0.8 GB
Over the limit 8.3 GB

Short by 8.3 GB. You can run it with 24 of 48 layers on the RTX 4070 Super and the rest in system RAM, at roughly 26.3 tok/s — usable for batch work, painful for chat. A smaller quantisation or a shorter context is usually the better trade.

Generation26.3tok/s
Prompt processing4464tok/s
Max context0tokens
KV per 1K tokens0GB

Every quantisation of Qwen3 30B-A3B on a RTX 4070 Super

Highlighted row is the highest quality that still fits at 8K context.

QuantisationbpwWeights TotalVerdictMax ctxtok/s
FP16 / BF16 16.00 56.9 GB 58.4 GB Won't fit — 6.45
INT8 / W8A8 8.50 30.1 GB 31.6 GB Won't fit — 13.2
Q8_0 (GGUF) 8.50 30.1 GB 31.6 GB Won't fit — 13.2
FP8 (E4M3) 8.00 28.4 GB 30.0 GB Won't fit — 14.1
Q6_K 6.56 23.3 GB 24.9 GB Won't fit — 17.8
Q5_K_M 5.67 20.2 GB 21.7 GB Won't fit — 21.6
Q5_K_S 5.52 19.6 GB 21.2 GB Won't fit — 22.0
Q4_K_M 4.85 17.3 GB 18.8 GB Won't fit — 26.3
Q4_K_S 4.58 16.3 GB 17.9 GB Won't fit — 29.0
Q4_0 4.55 16.2 GB 17.8 GB Won't fit — 29.1
AWQ 4-bit 4.25 16.0 GB 17.5 GB Won't fit — 29.5
GPTQ 4-bit 4.25 16.0 GB 17.5 GB Won't fit — 29.5
MXFP4 4.25 16.0 GB 17.5 GB Won't fit — 29.5
IQ4_XS 4.25 15.2 GB 16.7 GB Won't fit — 32.4
Q3_K_M 3.91 14.0 GB 15.5 GB Won't fit — 36.6
IQ3_M 3.70 13.3 GB 14.8 GB Won't fit — 40.5
IQ3_XXS 3.06 11.1 GB 12.6 GB Won't fit — 57.1
Q2_K 2.63 9.6 GB 11.1 GB Won't fit 1K 82.1
IQ2_XXS 2.06 7.6 GB 9.1 GB Runs comfortably 23K 115
IQ1_M 1.75 6.5 GB 8.0 GB Runs comfortably 34K 121

Also worth checking

Qwen3 30B-A3B on every GPU
Full hardware tableEverything that runs on a RTX 4070 Super
Full model tableWhy the KV cache is the part that bites
Guide