Qwen on NVIDIA Blackwell

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

Not at Q4_K_M — it needs 18.8 GB against 14.2 GB available. Drop to IQ3_XXS and it fits, at about 139 tokens per second.

Won't fit

18.8 GB of 14.2 GB · 132%
020 GB
Weights 17.3 GB
KV cache 0.8 GB
Runtime overhead 0.8 GB
Over the limit 4.6 GB

Short by 4.6 GB. You can run it with 35 of 48 layers on the RTX 5070 Ti and the rest in system RAM, at roughly 45.7 tok/s — usable for batch work, painful for chat. A smaller quantisation or a shorter context is usually the better trade.

Generation45.7tok/s
Prompt processing5533tok/s
Max context0tokens
KV per 1K tokens0GB

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

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 — 7.66
INT8 / W8A8 8.50 30.1 GB 31.6 GB Won't fit — 17.3
Q8_0 (GGUF) 8.50 30.1 GB 31.6 GB Won't fit — 17.3
FP8 (E4M3) 8.00 28.4 GB 30.0 GB Won't fit — 18.7
Q6_K 6.56 23.3 GB 24.9 GB Won't fit — 25.7
Q5_K_M 5.67 20.2 GB 21.7 GB Won't fit — 33.1
Q5_K_S 5.52 19.6 GB 21.2 GB Won't fit — 35.2
Q4_K_M 4.85 17.3 GB 18.8 GB Won't fit — 45.7
Q4_K_S 4.58 16.3 GB 17.9 GB Won't fit — 52.3
Q4_0 4.55 16.2 GB 17.8 GB Won't fit — 52.6
AWQ 4-bit 4.25 16.0 GB 17.5 GB Won't fit — 56.0
GPTQ 4-bit 4.25 16.0 GB 17.5 GB Won't fit — 56.0
MXFP4 4.25 16.0 GB 17.5 GB Won't fit — 56.0
IQ4_XS 4.25 15.2 GB 16.7 GB Won't fit — 64.7
Q3_K_M 3.91 14.0 GB 15.5 GB Won't fit — 82.3
IQ3_M 3.70 13.3 GB 14.8 GB Won't fit 2K 98.5
IQ3_XXS 3.06 11.1 GB 12.6 GB Runs comfortably 26K 139
Q2_K 2.63 9.6 GB 11.1 GB Runs comfortably 42K 146
IQ2_XXS 2.06 7.6 GB 9.1 GB Runs comfortably 63K 155
IQ1_M 1.75 6.5 GB 8.0 GB Runs comfortably 74K 161

Also worth checking

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