localLLMriglocal, on your table

estimate

Can I run Qwen2.5 14B on an RTX 5080?

Yes, fully in VRAM (estimate)

Can I run Qwen2.5 14B on an RTX 5080? Yes, fully in VRAM (estimate). About 11.2 GB at Q4_K_M (estimate).

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

Qwen2.5 14B on RTX 5080, formula estimate
QuantContextWeightsKVOverheadTotalUsableFitEst. tok/s
Q4 Q4_K_M8,192 tokens8.8 GB estimate1.6 GB estimate0.8 GB estimate11.2 GB estimate15.4 GB estimatefull71 tok/s estimate
Q5 Q5 (~0.71 bytes/param)8,192 tokens10.4 GB estimate1.6 GB estimate0.8 GB estimate12.8 GB estimate15.4 GB estimatefull61 tok/s estimate
Q8 Q88,192 tokens15.7 GB estimate1.6 GB estimate0.8 GB estimate18.1 GB estimate15.4 GB estimateno—
FP16 FP168,192 tokens29.4 GB estimate1.6 GB estimate0.8 GB estimate31.8 GB estimate15.4 GB estimateno—
Q4, longer context32,768 tokens8.8 GB estimate6.4 GB estimate0.8 GB estimate16.1 GB estimate15.4 GB estimateno—

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

$1,468€1,311

new (lowest in-stock EU)

price as of 17 Oct 2026

GPU PRIX

Lowest in-stock board in the table on 17 Oct 2026: €1,311, ASUS Prime GeForce RTX 5080, in stock. The same page's header current low was €1,328, last updated Oct 17, 2026. 16 GB GDDR7 and 360 W total graphics power are on NVIDIA's RTX 5080 spec page (https://www.nvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5080/). 960 GB/sec is the RTX 5080 column on NVIDIA's GeForce compare page (https://www.nvidia.com/en-us/geforce/graphics-cards/compare/). The product spec table that was fetched did not print a GB/s row.

Embed this

Qwen2.5 14B on RTX 5080

[![Qwen2.5 14B Q4 on RTX 5080: fits ✓ · est. 71 tok/s (estimate)](https://localllmrig.com/badge/qwen2.5-14b/rtx-5080.svg)](https://localllmrig.com/can-i-run/qwen2.5-14b/rtx-5080?utm_source=badge)

Questions

Can I run Qwen2.5 14B on an RTX 5080?
Yes, fully in VRAM (estimate).
How much memory does Qwen2.5 14B need at Q4?
About 11.2 GB at 8,192 tokens (estimate): 8.8 GB weights, 1.6 GB KV cache, 0.8 GB overhead.
How fast is Qwen2.5 14B on RTX 5080 (estimate)?
About 71 tok/s at Q4 (estimate, not a benchmark).
Qwen2.5 14B on RTX 5080 (estimate) · localLLMrig