// RUN ON CONSUMER GPU

Run Polyglot-Ko-12.8B on Consumer GPU hardware

28 options ranked by estimated speed at Q4 / 8k context.

HardwareEst tok/sFits 8k?Max ctx~Price
NVIDIA GeForce RTX 5090 → ~215.8 yes 126k ~$1,999
NVIDIA GeForce RTX 4090 → ~121.4 yes 82k ~$1,599
NVIDIA GeForce RTX 3090 Ti → ~121.4 yes 82k ~$1,999
NVIDIA GeForce RTX 5080 → ~115.6 yes 37k ~$999
NVIDIA GeForce RTX 3090 → ~112.7 yes 82k ~$1,499
NVIDIA GeForce RTX 3080 Ti → ~109.8 yes 14k ~$1,199
NVIDIA GeForce RTX 5070 Ti → ~107.9 yes 37k ~$749
AMD Radeon RX 7900 XTX → ~102.3 yes 82k ~$999
NVIDIA GeForce RTX 3080 10GB → ~91.5 tight 3k ~$699
NVIDIA GeForce RTX 4080 → ~88.6 yes 37k ~$1,199
AMD Radeon RX 7900 XT → ~85.3 yes 59k ~$899
NVIDIA GeForce RTX 5070 → ~80.9 yes 14k ~$549
NVIDIA GeForce RTX 4070 Ti Super → ~80.9 yes 37k ~$799
AMD Radeon RX 9070 XT → ~68.7 yes 37k ~$599
AMD Radeon RX 7800 XT → ~66.5 yes 37k ~$499
NVIDIA GeForce RTX 4070 Super → ~60.7 yes 14k ~$599
NVIDIA GeForce RTX 4070 → ~60.7 yes 14k ~$599
Intel Arc Pro B70 → ~60.6 yes 126k —
Intel Arc A770 16GB → ~55.8 yes 37k ~$329
AMD Radeon RX 6800 XT → ~54.6 yes 37k ~$649
AMD Radeon RX 6900 XT → ~54.6 yes 37k ~$999
NVIDIA GeForce RTX 5060 Ti 16GB → ~54 yes 37k ~$429
Intel Arc Pro B60 → ~45.5 yes 82k —
Intel Arc B580 12GB → ~45.5 yes 14k ~$249
NVIDIA GeForce RTX 4060 Ti 16GB → ~34.7 yes 37k ~$499
NVIDIA DGX Spark → ~32.9 yes 650k ~$3,999
AMD Ryzen AI Max+ 395 → ~27.3 yes 650k ~$1,999
Intel Arc Pro B50 → ~22.3 yes 37k —

Estimate at Q4 (memory-bound). Click any option for its full per-hardware estimate and measured numbers.

→ Polyglot-Ko-12.8B overview → All hardware

FAQ

What Consumer GPU hardware runs Polyglot-Ko-12.8B?▶

On GenAIList, 28 Consumer GPU options can run Polyglot-Ko-12.8B at Q4 — the fastest is the NVIDIA GeForce RTX 5090 at ~215.8 tok/s.

How fast is Polyglot-Ko-12.8B on Consumer GPU hardware?▶

Estimated generation speed ranges from ~22.3 to ~215.8 tok/s at Q4 (memory-bandwidth bound). Click any option for its full estimate and any measured numbers.