localLLMriglocal, on your table

estimate

Can I run Llama 3.2 3B on an RTX 3060 12GB?

Yes, fully in VRAM (estimate)

Can I run Llama 3.2 3B on an RTX 3060 12GB? Yes, fully in VRAM (estimate). About 3.7 GB at Q4_K_M (estimate).

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

Llama 3.2 3B on RTX 3060 12GB, formula estimate
QuantContextWeightsKVOverheadTotalUsableFitEst. tok/s
Q4 Q4_K_M8,192 tokens1.9 GB estimate0.9 GB estimate0.8 GB estimate3.7 GB estimate11.4 GB estimatefull96 tok/s estimate
Q5 Q5 (~0.71 bytes/param)8,192 tokens2.3 GB estimate0.9 GB estimate0.8 GB estimate4.0 GB estimate11.4 GB estimatefull86 tok/s estimate
Q8 Q88,192 tokens3.4 GB estimate0.9 GB estimate0.8 GB estimate5.2 GB estimate11.4 GB estimatefull63 tok/s estimate
FP16 FP168,192 tokens6.4 GB estimate0.9 GB estimate0.8 GB estimate8.2 GB estimate11.4 GB estimatefull38 tok/s estimate
Q4, longer context32,768 tokens1.9 GB estimate3.8 GB estimate0.8 GB estimate6.5 GB estimate11.4 GB estimatefull49 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

$288€257

used (median of sold listings)

price as of 16 Oct 2026

pcprice.watch

Used median €257, typical €236–€280, 1,378 sales. The buying-guide page says Updated Oct 15, 2026. Recorded here on 16 Oct 2026. Not an EU retail listing. No separate new in-stock figure is stored for this card.

Embed this

Llama 3.2 3B on RTX 3060 12GB

[![Llama 3.2 3B Q4 on RTX 3060 12GB: fits ✓ · est. 96 tok/s (estimate)](https://localllmrig.com/badge/llama-3.2-3b/rtx-3060-12gb.svg)](https://localllmrig.com/can-i-run/llama-3.2-3b/rtx-3060-12gb?utm_source=badge)

Questions

Can I run Llama 3.2 3B on an RTX 3060 12GB?
Yes, fully in VRAM (estimate).
How much memory does Llama 3.2 3B need at Q4?
About 3.7 GB at 8,192 tokens (estimate): 1.9 GB weights, 0.9 GB KV cache, 0.8 GB overhead.
How fast is Llama 3.2 3B on RTX 3060 12GB (estimate)?
About 96 tok/s at Q4 (estimate, not a benchmark).
Llama 3.2 3B on RTX 3060 12GB (estimate) · localLLMrig