// RUN ON CONSUMER GPU

Run phi-3.5-mini 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 → ~732.6 yes 394k ~$1,999
NVIDIA GeForce RTX 4090 → ~412.1 yes 282k ~$1,599
NVIDIA GeForce RTX 3090 Ti → ~412.1 yes 282k ~$1,999
NVIDIA GeForce RTX 5080 → ~392.5 yes 169k ~$999
NVIDIA GeForce RTX 3090 → ~382.7 yes 282k ~$1,499
NVIDIA GeForce RTX 3080 Ti → ~372.9 yes 113k ~$1,199
NVIDIA GeForce RTX 5070 Ti → ~366.3 yes 169k ~$749
AMD Radeon RX 7900 XTX → ~347.4 yes 282k ~$999
NVIDIA GeForce RTX 3080 10GB → ~310.7 yes 85k ~$699
NVIDIA GeForce RTX 4080 → ~300.9 yes 169k ~$1,199
AMD Radeon RX 7900 XT → ~289.5 yes 225k ~$899
NVIDIA GeForce RTX 5070 → ~274.7 yes 113k ~$549
NVIDIA GeForce RTX 4070 Ti Super → ~274.7 yes 169k ~$799
NVIDIA GeForce RTX 3070 Ti → ~248.6 yes 57k ~$599
AMD Radeon RX 9070 XT → ~233 yes 169k ~$599
AMD Radeon RX 7800 XT → ~225.8 yes 169k ~$499
NVIDIA GeForce RTX 4070 Super → ~206.1 yes 113k ~$599
NVIDIA GeForce RTX 4070 → ~206.1 yes 113k ~$599
Intel Arc Pro B70 → ~205.7 yes 394k —
Intel Arc A770 16GB → ~189.5 yes 169k ~$329
AMD Radeon RX 6800 XT → ~185.3 yes 169k ~$649
AMD Radeon RX 6900 XT → ~185.3 yes 169k ~$999
NVIDIA GeForce RTX 3070 → ~183.2 yes 57k ~$499
NVIDIA GeForce RTX 5060 Ti 16GB → ~183.2 yes 169k ~$429
NVIDIA GeForce RTX 5060 Ti 8GB → ~183.2 yes 57k ~$379
Intel Arc Pro B60 → ~154.3 yes 282k —
Intel Arc B580 12GB → ~154.3 yes 113k ~$249
NVIDIA GeForce RTX 4060 Ti 8GB → ~117.7 yes 57k ~$399
NVIDIA GeForce RTX 4060 Ti 16GB → ~117.7 yes 169k ~$499
NVIDIA DGX Spark → ~111.6 yes 1702k ~$3,999
AMD Ryzen AI Max+ 395 → ~92.6 yes 1702k ~$1,999
Intel Arc Pro B50 → ~75.8 yes 169k —

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

→ phi-3.5-mini overview → All hardware

FAQ

What Consumer GPU hardware runs phi-3.5-mini?▶

On GenAIList, 32 Consumer GPU options can run phi-3.5-mini at Q4 — the fastest is the NVIDIA GeForce RTX 5090 at ~732.6 tok/s.

How fast is phi-3.5-mini on Consumer GPU hardware?▶

Estimated generation speed ranges from ~75.8 to ~732.6 tok/s at Q4 (memory-bandwidth bound). Click any option for its full estimate and any measured numbers.