// RUN ON CONSUMER GPU

Run Yi-1.5-9B on Consumer GPU hardware

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

HardwareEst tok/sFits 8k?Max ctx~Price
NVIDIA GeForce RTX 5090 → ~315.3 yes 175k ~$1,999
NVIDIA GeForce RTX 4090 → ~177.3 yes 119k ~$1,599
NVIDIA GeForce RTX 3090 Ti → ~177.3 yes 119k ~$1,999
NVIDIA GeForce RTX 5080 → ~168.9 yes 63k ~$999
NVIDIA GeForce RTX 3090 → ~164.7 yes 119k ~$1,499
NVIDIA GeForce RTX 3080 Ti → ~160.5 yes 35k ~$1,199
NVIDIA GeForce RTX 5070 Ti → ~157.6 yes 63k ~$749
AMD Radeon RX 7900 XTX → ~149.5 yes 119k ~$999
NVIDIA GeForce RTX 3080 10GB → ~133.7 yes 21k ~$699
NVIDIA GeForce RTX 4080 → ~129.5 yes 63k ~$1,199
AMD Radeon RX 7900 XT → ~124.6 yes 91k ~$899
NVIDIA GeForce RTX 5070 → ~118.2 yes 35k ~$549
NVIDIA GeForce RTX 4070 Ti Super → ~118.2 yes 63k ~$799
NVIDIA GeForce RTX 3070 Ti → ~107 tight 7k ~$599
AMD Radeon RX 9070 XT → ~100.3 yes 63k ~$599
AMD Radeon RX 7800 XT → ~97.2 yes 63k ~$499
NVIDIA GeForce RTX 4070 Super → ~88.7 yes 35k ~$599
NVIDIA GeForce RTX 4070 → ~88.7 yes 35k ~$599
Intel Arc Pro B70 → ~88.5 yes 175k —
Intel Arc A770 16GB → ~81.5 yes 63k ~$329
AMD Radeon RX 6800 XT → ~79.7 yes 63k ~$649
AMD Radeon RX 6900 XT → ~79.7 yes 63k ~$999
NVIDIA GeForce RTX 3070 → ~78.8 tight 7k ~$499
NVIDIA GeForce RTX 5060 Ti 16GB → ~78.8 yes 63k ~$429
NVIDIA GeForce RTX 5060 Ti 8GB → ~78.8 tight 7k ~$379
Intel Arc Pro B60 → ~66.4 yes 119k —
Intel Arc B580 12GB → ~66.4 yes 35k ~$249
NVIDIA GeForce RTX 4060 Ti 8GB → ~50.7 tight 7k ~$399
NVIDIA GeForce RTX 4060 Ti 16GB → ~50.7 yes 63k ~$499
NVIDIA DGX Spark → ~48 yes 830k ~$3,999
AMD Ryzen AI Max+ 395 → ~39.9 yes 830k ~$1,999
Intel Arc Pro B50 → ~32.6 yes 63k —

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

→ Yi-1.5-9B overview → All hardware

FAQ

What Consumer GPU hardware runs Yi-1.5-9B?▶

On GenAIList, 32 Consumer GPU options can run Yi-1.5-9B at Q4 — the fastest is the NVIDIA GeForce RTX 5090 at ~315.3 tok/s.

How fast is Yi-1.5-9B on Consumer GPU hardware?▶

Estimated generation speed ranges from ~32.6 to ~315.3 tok/s at Q4 (memory-bandwidth bound). Click any option for its full estimate and any measured numbers.