Qwen on NVIDIA Ada

Can I run Qwen3 8B on an RTX 4060 Ti 16GB?

Yes. Qwen3 8B at Q4_K_M uses 6.7 GB of the 14.2 GB available on a RTX 4060 Ti 16GB, and runs at about 32.5 tokens per second. You can push the context to 62K.

Runs comfortably

6.7 GB of 14.2 GB · 47%
014 GB
Weights 4.7 GB
KV cache 1.1 GB
Runtime overhead 0.8 GB

Qwen3 8B at Q4_K_M leaves 7.6 GB spare on a RTX 4060 Ti 16GB. There is room to raise the context length or move up a quantisation level.

Generation32.5tok/s
Prompt processing1128tok/s
Max context62Ktokens
KV per 1K tokens0GB

Every quantisation of Qwen3 8B 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 15.3 GB 17.2 GB Won't fit — 6.48
INT8 / W8A8 8.50 7.8 GB 9.8 GB Runs comfortably 40K 20.5
Q8_0 (GGUF) 8.50 7.8 GB 9.8 GB Runs comfortably 40K 20.5
FP8 (E4M3) 8.00 7.6 GB 9.6 GB Runs comfortably 41K 21.0
Q6_K 6.56 6.3 GB 8.2 GB Runs comfortably 51K 25.2
AWQ 4-bit 4.25 5.8 GB 7.7 GB Runs comfortably 55K 27.2
GPTQ 4-bit 4.25 5.8 GB 7.7 GB Runs comfortably 55K 27.2
MXFP4 4.25 5.8 GB 7.7 GB Runs comfortably 55K 27.2
Q5_K_M 5.67 5.4 GB 7.4 GB Runs comfortably 57K 28.8
Q5_K_S 5.52 5.3 GB 7.2 GB Runs comfortably 58K 29.5
Q4_K_M 4.85 4.7 GB 6.7 GB Runs comfortably 62K 32.5
Q4_K_S 4.58 4.5 GB 6.4 GB Runs comfortably 63K 33.9
Q4_0 4.55 4.5 GB 6.4 GB Runs comfortably 64K 34.0
IQ4_XS 4.25 4.2 GB 6.2 GB Runs comfortably 65K 35.7
Q3_K_M 3.91 4.0 GB 5.9 GB Runs comfortably 67K 37.9
IQ3_M 3.70 3.8 GB 5.7 GB Runs comfortably 68K 39.4
IQ3_XXS 3.06 3.3 GB 5.2 GB Runs comfortably 72K 44.6
Q2_K 2.63 2.9 GB 4.9 GB Runs comfortably 75K 49.0
IQ2_XXS 2.06 2.5 GB 4.4 GB Runs comfortably 78K 56.4
IQ1_M 1.75 2.2 GB 4.2 GB Runs comfortably 80K 61.4

Also worth checking

Qwen3 8B 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