Qwen on NVIDIA Blackwell

Can I run Qwen3 30B-A3B on an RTX 5070?

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 5070 and the rest in system RAM, at roughly 29.5 tok/s — usable for batch work, painful for chat. A smaller quantisation or a shorter context is usually the better trade.

Generation29.5tok/s
Prompt processing3898tok/s
Max context0tokens
KV per 1K tokens0GB

Every quantisation of Qwen3 30B-A3B on a RTX 5070

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 — 7.14
INT8 / W8A8 8.50 30.1 GB 31.6 GB Won't fit — 14.6
Q8_0 (GGUF) 8.50 30.1 GB 31.6 GB Won't fit — 14.6
FP8 (E4M3) 8.00 28.4 GB 30.0 GB Won't fit — 15.7
Q6_K 6.56 23.3 GB 24.9 GB Won't fit — 19.9
Q5_K_M 5.67 20.2 GB 21.7 GB Won't fit — 24.1
Q5_K_S 5.52 19.6 GB 21.2 GB Won't fit — 24.6
Q4_K_M 4.85 17.3 GB 18.8 GB Won't fit — 29.5
Q4_K_S 4.58 16.3 GB 17.9 GB Won't fit — 32.6
Q4_0 4.55 16.2 GB 17.8 GB Won't fit — 32.8
AWQ 4-bit 4.25 16.0 GB 17.5 GB Won't fit — 33.2
GPTQ 4-bit 4.25 16.0 GB 17.5 GB Won't fit — 33.2
MXFP4 4.25 16.0 GB 17.5 GB Won't fit — 33.2
IQ4_XS 4.25 15.2 GB 16.7 GB Won't fit — 36.6
Q3_K_M 3.91 14.0 GB 15.5 GB Won't fit — 41.4
IQ3_M 3.70 13.3 GB 14.8 GB Won't fit — 46.1
IQ3_XXS 3.06 11.1 GB 12.6 GB Won't fit — 65.9
Q2_K 2.63 9.6 GB 11.1 GB Won't fit 1K 97.1
IQ2_XXS 2.06 7.6 GB 9.1 GB Runs comfortably 23K 138
IQ1_M 1.75 6.5 GB 8.0 GB Runs comfortably 34K 144

Also worth checking

Qwen3 30B-A3B on every GPU
Full hardware tableEverything that runs on a RTX 5070
Full model tableWhy the KV cache is the part that bites
Guide