Qwen on NVIDIA Ada

Can I run Qwen3 30B-A3B on an RTX 4060 Ti 16GB?

Not at Q4_K_M — it needs 18.8 GB against 14.2 GB available. Drop to IQ3_XXS and it fits, at about 67.8 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 4060 Ti 16GB and the rest in system RAM, at roughly 32.0 tok/s — usable for batch work, painful for chat. A smaller quantisation or a shorter context is usually the better trade.

Generation32.0tok/s
Prompt processing2766tok/s
Max context0tokens
KV per 1K tokens0GB

Every quantisation of Qwen3 30B-A3B on a RTX 4060 Ti 16GB

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 6.68
INT8 / W8A8 8.50 30.1 GB 31.6 GB Won't fit 14.2
Q8_0 (GGUF) 8.50 30.1 GB 31.6 GB Won't fit 14.2
FP8 (E4M3) 8.00 28.4 GB 30.0 GB Won't fit 15.3
Q6_K 6.56 23.3 GB 24.9 GB Won't fit 20.1
Q5_K_M 5.67 20.2 GB 21.7 GB Won't fit 24.9
Q5_K_S 5.52 19.6 GB 21.2 GB Won't fit 26.1
Q4_K_M 4.85 17.3 GB 18.8 GB Won't fit 32.0
Q4_K_S 4.58 16.3 GB 17.9 GB Won't fit 35.4
Q4_0 4.55 16.2 GB 17.8 GB Won't fit 35.6
AWQ 4-bit 4.25 16.0 GB 17.5 GB Won't fit 37.1
GPTQ 4-bit 4.25 16.0 GB 17.5 GB Won't fit 37.1
MXFP4 4.25 16.0 GB 17.5 GB Won't fit 37.1
IQ4_XS 4.25 15.2 GB 16.7 GB Won't fit 41.0
Q3_K_M 3.91 14.0 GB 15.5 GB Won't fit 48.1
IQ3_M 3.70 13.3 GB 14.8 GB Won't fit 2K 53.8
IQ3_XXS 3.06 11.1 GB 12.6 GB Runs comfortably 26K 67.8
Q2_K 2.63 9.6 GB 11.1 GB Runs comfortably 42K 73.3
IQ2_XXS 2.06 7.6 GB 9.1 GB Runs comfortably 63K 82.1
IQ1_M 1.75 6.5 GB 8.0 GB Runs comfortably 74K 87.8

Also worth checking