Gemma on NVIDIA Ampere

Can I run Gemma 3 12B on an RTX A6000?

Yes. Gemma 3 12B at Q4_K_M uses 8.6 GB of the 44.3 GB available on a RTX A6000, and runs at about 62.7 tokens per second. You can push the context to 128K.

Runs comfortably

8.6 GB of 44.3 GB · 19%
044 GB
Weights 7.0 GB
KV cache 0.8 GB
Runtime overhead 0.8 GB

Gemma 3 12B at Q4_K_M leaves 35.7 GB spare on a RTX A6000. There is room to raise the context length or move up a quantisation level.

Generation62.7tok/s
Prompt processing1291tok/s
Max context128Ktokens
KV per 1K tokens0GB

Every quantisation of Gemma 3 12B on a RTX A6000

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 Runs comfortably 128K 20.5
INT8 / W8A8 8.50 11.8 GB 13.5 GB Runs comfortably 128K 38.3
Q8_0 (GGUF) 8.50 11.8 GB 13.5 GB Runs comfortably 128K 38.3
FP8 (E4M3) 8.00 11.4 GB 13.0 GB Runs comfortably 128K 39.9
Q6_K 6.56 9.3 GB 10.9 GB Runs comfortably 128K 48.0
Q5_K_M 5.67 8.1 GB 9.7 GB Runs comfortably 128K 54.9
Q5_K_S 5.52 7.8 GB 9.5 GB Runs comfortably 128K 56.3
AWQ 4-bit 4.25 7.4 GB 9.0 GB Runs comfortably 128K 59.2
GPTQ 4-bit 4.25 7.4 GB 9.0 GB Runs comfortably 128K 59.2
MXFP4 4.25 7.4 GB 9.0 GB Runs comfortably 128K 59.2
Q4_K_M 4.85 7.0 GB 8.6 GB Runs comfortably 128K 62.7
Q4_K_S 4.58 6.6 GB 8.2 GB Runs comfortably 128K 65.7
Q4_0 4.55 6.6 GB 8.2 GB Runs comfortably 128K 66.1
IQ4_XS 4.25 6.2 GB 7.8 GB Runs comfortably 128K 69.8
Q3_K_M 3.91 5.7 GB 7.4 GB Runs comfortably 128K 74.6
IQ3_M 3.70 5.5 GB 7.1 GB Runs comfortably 128K 78.0
IQ3_XXS 3.06 4.6 GB 6.3 GB Runs comfortably 128K 90.1
Q2_K 2.63 4.1 GB 5.7 GB Runs comfortably 128K 101
IQ2_XXS 2.06 3.3 GB 5.0 GB Runs comfortably 128K 119
IQ1_M 1.75 2.9 GB 4.6 GB Runs comfortably 128K 133

Also worth checking

Gemma 3 12B on every GPU
Full hardware tableEverything that runs on a RTX A6000
Full model tableWhy the KV cache is the part that bites
Guide