Qwen on NVIDIA Ampere

Can I run Qwen3 30B-A3B on an RTX 3080 10GB?

Not at Q4_K_M — it needs 18.8 GB against 8.6 GB available. You would need 3 of these cards.

Won't fit

18.8 GB of 8.6 GB · 219%
020 GB
Weights 17.3 GB
KV cache 0.8 GB
Runtime overhead 0.8 GB
Over the limit 10.2 GB

Short by 10.2 GB. You can run it with 19 of 48 layers on the RTX 3080 10GB and the rest in system RAM, at roughly 24.7 tok/s — usable for batch work, painful for chat. A smaller quantisation or a shorter context is usually the better trade.

Generation24.7tok/s
Prompt processing3772tok/s
Max context0tokens
KV per 1K tokens0GB

Every quantisation of Qwen3 30B-A3B on a RTX 3080 10GB

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 — 6.51
INT8 / W8A8 8.50 30.1 GB 31.6 GB Won't fit — 13.0
Q8_0 (GGUF) 8.50 30.1 GB 31.6 GB Won't fit — 13.0
FP8 (E4M3) 8.00 28.4 GB 30.0 GB Won't fit — 13.6
Q6_K 6.56 23.3 GB 24.9 GB Won't fit — 17.2
Q5_K_M 5.67 20.2 GB 21.7 GB Won't fit — 20.3
Q5_K_S 5.52 19.6 GB 21.2 GB Won't fit — 21.3
Q4_K_M 4.85 17.3 GB 18.8 GB Won't fit — 24.7
Q4_K_S 4.58 16.3 GB 17.9 GB Won't fit — 26.4
Q4_0 4.55 16.2 GB 17.8 GB Won't fit — 26.5
AWQ 4-bit 4.25 16.0 GB 17.5 GB Won't fit — 27.6
GPTQ 4-bit 4.25 16.0 GB 17.5 GB Won't fit — 27.6
MXFP4 4.25 16.0 GB 17.5 GB Won't fit — 27.6
IQ4_XS 4.25 15.2 GB 16.7 GB Won't fit — 29.4
Q3_K_M 3.91 14.0 GB 15.5 GB Won't fit — 33.0
IQ3_M 3.70 13.3 GB 14.8 GB Won't fit — 35.3
IQ3_XXS 3.06 11.1 GB 12.6 GB Won't fit — 46.8
Q2_K 2.63 9.6 GB 11.1 GB Won't fit — 61.6
IQ2_XXS 2.06 7.6 GB 9.1 GB Won't fit 3K 108
IQ1_M 1.75 6.5 GB 8.0 GB Fits, but tight 14K 149

Also worth checking

Qwen3 30B-A3B on every GPU
Full hardware tableEverything that runs on a RTX 3080 10GB
Full model tableWhy the KV cache is the part that bites
Guide