localLLMriglocal, on your table

estimate

Can I run Gemma 2 9B on an RTX 3090?

Yes, fully in VRAM (estimate)

Can I run Gemma 2 9B on an RTX 3090? Yes, fully in VRAM (estimate). About 8.5 GB at Q4_K_M (estimate).

Updated . The later of the hardware price date and the formula date.

Gemma 2 9B on RTX 3090, formula estimate
QuantContextWeightsKVOverheadTotalUsableFitEst. tok/s
Q4 Q4_K_M8,192 tokens5.5 GB estimate2.1 GB estimate0.8 GB estimate8.5 GB estimate23.4 GB estimatefull94 tok/s estimate
Q5 Q5 (~0.71 bytes/param)8,192 tokens6.6 GB estimate2.1 GB estimate0.8 GB estimate9.5 GB estimate23.4 GB estimatefull83 tok/s estimate
Q8 Q88,192 tokens9.9 GB estimate2.1 GB estimate0.8 GB estimate12.8 GB estimate23.4 GB estimatefull60 tok/s estimate
FP16 FP168,192 tokens18.5 GB estimate2.1 GB estimate0.8 GB estimate21.4 GB estimate23.4 GB estimatefull35 tok/s estimate
Q4, longer context8,192 tokens5.5 GB estimate2.1 GB estimate0.8 GB estimate8.5 GB estimate23.4 GB estimatefull94 tok/s estimate

Formula estimate, not a benchmark. Method →

Cheapest catalog option that fits at Q4

GeForce RTX 3060 12GB, used (median of sold listings), €257, dated 2026-10-16. No buy link.

This hardware

$1,370€1,224

used (median of sold listings)

price as of 16 Oct 2026

new (lowest in-stock EU): $2,483€2,218. GPU PRIX. New low €2,218, Zotac Trinity OC, France. Page last updated 16 Oct 2026.

pcprice.watch

Used median €1,224, typical €1,136–€1,312, 232 eBay sold sales over 90 days, updated 16 Oct 2026. Not an EU retail listing.

Embed this

Gemma 2 9B on RTX 3090

[![Gemma 2 9B Q4 on RTX 3090: fits ✓ · est. 94 tok/s (estimate)](https://localllmrig.com/badge/gemma-2-9b/rtx-3090.svg)](https://localllmrig.com/can-i-run/gemma-2-9b/rtx-3090?utm_source=badge)

Questions

Can I run Gemma 2 9B on an RTX 3090?
Yes, fully in VRAM (estimate).
How much memory does Gemma 2 9B need at Q4?
About 8.5 GB at 8,192 tokens (estimate): 5.5 GB weights, 2.1 GB KV cache, 0.8 GB overhead.
How fast is Gemma 2 9B on RTX 3090 (estimate)?
About 94 tok/s at Q4 (estimate, not a benchmark).
Can I run Gemma 2 9B on an RTX 3090? VRAM & speed (estimate) · localLLMrig