Qwen on NVIDIA Ada

Can I run Qwen3 14B on an RTX 4060 Ti 16GB?

Yes. Qwen3 14B at Q4_K_M uses 10.5 GB of the 14.2 GB available on a RTX 4060 Ti 16GB, and runs at about 18.9 tokens per second. You can push the context to 32K.

Runs comfortably

10.5 GB of 14.2 GB · 74%
014 GB
Weights 8.5 GB
KV cache 1.3 GB
Runtime overhead 0.8 GB

Qwen3 14B at Q4_K_M leaves 3.7 GB spare on a RTX 4060 Ti 16GB. There is room to raise the context length or move up a quantisation level.

Generation18.9tok/s
Prompt processing626tok/s
Max context32Ktokens
KV per 1K tokens0GB

Every quantisation of Qwen3 14B 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 27.5 GB 29.6 GB Won't fit — 2.03
INT8 / W8A8 8.50 14.3 GB 16.4 GB Won't fit — 7.58
Q8_0 (GGUF) 8.50 14.3 GB 16.4 GB Won't fit — 7.58
FP8 (E4M3) 8.00 13.8 GB 15.8 GB Won't fit — 8.33
Q6_K 6.56 11.3 GB 13.4 GB Fits, but tight 14K 14.5
Q5_K_M 5.67 9.7 GB 11.8 GB Runs comfortably 23K 16.6
Q5_K_S 5.52 9.5 GB 11.6 GB Runs comfortably 25K 17.0
AWQ 4-bit 4.25 9.4 GB 11.5 GB Runs comfortably 25K 17.1
GPTQ 4-bit 4.25 9.4 GB 11.5 GB Runs comfortably 25K 17.1
MXFP4 4.25 9.4 GB 11.5 GB Runs comfortably 25K 17.1
Q4_K_M 4.85 8.5 GB 10.5 GB Runs comfortably 32K 18.9
Q4_K_S 4.58 8.0 GB 10.1 GB Runs comfortably 34K 19.8
Q4_0 4.55 8.0 GB 10.1 GB Runs comfortably 35K 19.9
IQ4_XS 4.25 7.5 GB 9.6 GB Runs comfortably 38K 21.1
Q3_K_M 3.91 7.0 GB 9.1 GB Runs comfortably 41K 22.5
IQ3_M 3.70 6.7 GB 8.8 GB Runs comfortably 43K 23.5
IQ3_XXS 3.06 5.7 GB 7.8 GB Runs comfortably 49K 27.1
Q2_K 2.63 5.0 GB 7.1 GB Runs comfortably 54K 30.3
IQ2_XXS 2.06 4.2 GB 6.3 GB Runs comfortably 59K 35.7
IQ1_M 1.75 3.7 GB 5.8 GB Runs comfortably 62K 39.7

Also worth checking

Qwen3 14B on every GPU
Full hardware tableEverything that runs on a RTX 4060 Ti 16GB
Full model tableWhy the KV cache is the part that bites
Guide