Gemma on NVIDIA Ampere

Can I run Gemma 3 12B on an RTX 3090?

Yes. Gemma 3 12B at Q4_K_M uses 8.6 GB of the 21.8 GB available on a RTX 3090, and runs at about 76.1 tokens per second. You can push the context to 128K.

Runs comfortably

8.6 GB of 21.8 GB · 39%
022 GB
Weights 7.0 GB
KV cache 0.8 GB
Runtime overhead 0.8 GB

Gemma 3 12B at Q4_K_M leaves 13.2 GB spare on a RTX 3090. There is room to raise the context length or move up a quantisation level.

Generation76.1tok/s
Prompt processing1222tok/s
Max context128Ktokens
KV per 1K tokens0GB

Every quantisation of Gemma 3 12B on a RTX 3090

Highlighted row is the highest quality that still fits at 8K context.

QuantisationbpwWeights TotalVerdictMax ctxtok/s
FP16 / BF16 16.00 22.7 GB 24.4 GB Won't fit — 9.27
INT8 / W8A8 8.50 11.8 GB 13.5 GB Runs comfortably 128K 46.6
Q8_0 (GGUF) 8.50 11.8 GB 13.5 GB Runs comfortably 128K 46.6
FP8 (E4M3) 8.00 11.4 GB 13.0 GB Runs comfortably 128K 48.4
Q6_K 6.56 9.3 GB 10.9 GB Runs comfortably 128K 58.3
Q5_K_M 5.67 8.1 GB 9.7 GB Runs comfortably 128K 66.7
Q5_K_S 5.52 7.8 GB 9.5 GB Runs comfortably 128K 68.4
AWQ 4-bit 4.25 7.4 GB 9.0 GB Runs comfortably 128K 71.9
GPTQ 4-bit 4.25 7.4 GB 9.0 GB Runs comfortably 128K 71.9
MXFP4 4.25 7.4 GB 9.0 GB Runs comfortably 128K 71.9
Q4_K_M 4.85 7.0 GB 8.6 GB Runs comfortably 128K 76.1
Q4_K_S 4.58 6.6 GB 8.2 GB Runs comfortably 128K 79.8
Q4_0 4.55 6.6 GB 8.2 GB Runs comfortably 128K 80.2
IQ4_XS 4.25 6.2 GB 7.8 GB Runs comfortably 128K 84.7
Q3_K_M 3.91 5.7 GB 7.4 GB Runs comfortably 128K 90.5
IQ3_M 3.70 5.5 GB 7.1 GB Runs comfortably 128K 94.5
IQ3_XXS 3.06 4.6 GB 6.3 GB Runs comfortably 128K 109
Q2_K 2.63 4.1 GB 5.7 GB Runs comfortably 128K 122
IQ2_XXS 2.06 3.3 GB 5.0 GB Runs comfortably 128K 144
IQ1_M 1.75 2.9 GB 4.6 GB Runs comfortably 128K 160

Also worth checking

Gemma 3 12B on every GPU
Full hardware tableEverything that runs on a RTX 3090
Full model tableWhy the KV cache is the part that bites
Guide