// RUN ON CONSUMER GPU

Run VILA-13B on Consumer GPU hardware

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

HardwareEst tok/sFits 8k?Max ctx~Price
NVIDIA GeForce RTX 5090 → ~208.5 yes 125k ~$1,999
NVIDIA GeForce RTX 4090 → ~117.3 yes 80k ~$1,599
NVIDIA GeForce RTX 3090 Ti → ~117.3 yes 80k ~$1,999
NVIDIA GeForce RTX 5080 → ~111.7 yes 35k ~$999
NVIDIA GeForce RTX 3090 → ~108.9 yes 80k ~$1,499
NVIDIA GeForce RTX 3080 Ti → ~106.1 yes 13k ~$1,199
NVIDIA GeForce RTX 5070 Ti → ~104.3 yes 35k ~$749
AMD Radeon RX 7900 XTX → ~98.9 yes 80k ~$999
NVIDIA GeForce RTX 4080 → ~85.6 yes 35k ~$1,199
AMD Radeon RX 7900 XT → ~82.4 yes 58k ~$899
NVIDIA GeForce RTX 5070 → ~78.2 yes 13k ~$549
NVIDIA GeForce RTX 4070 Ti Super → ~78.2 yes 35k ~$799
AMD Radeon RX 9070 XT → ~66.3 yes 35k ~$599
AMD Radeon RX 7800 XT → ~64.3 yes 35k ~$499
Intel Arc Pro B70 → ~58.6 yes 125k —
NVIDIA GeForce RTX 4070 Super → ~58.6 yes 13k ~$599
NVIDIA GeForce RTX 4070 → ~58.6 yes 13k ~$599
Intel Arc A770 16GB → ~53.9 yes 35k ~$329
AMD Radeon RX 6800 XT → ~52.7 yes 35k ~$649
AMD Radeon RX 6900 XT → ~52.7 yes 35k ~$999
NVIDIA GeForce RTX 5060 Ti 16GB → ~52.1 yes 35k ~$429
Intel Arc Pro B60 → ~43.9 yes 80k —
Intel Arc B580 12GB → ~43.9 yes 13k ~$249
NVIDIA GeForce RTX 4060 Ti 16GB → ~33.5 yes 35k ~$499
NVIDIA DGX Spark → ~31.8 yes 648k ~$3,999
AMD Ryzen AI Max+ 395 → ~26.4 yes 648k ~$1,999
Intel Arc Pro B50 → ~21.6 yes 35k —

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

→ VILA-13B overview → All hardware

FAQ

What Consumer GPU hardware runs VILA-13B?▶

On GenAIList, 27 Consumer GPU options can run VILA-13B at Q4 — the fastest is the NVIDIA GeForce RTX 5090 at ~208.5 tok/s.

How fast is VILA-13B on Consumer GPU hardware?▶

Estimated generation speed ranges from ~21.6 to ~208.5 tok/s at Q4 (memory-bandwidth bound). Click any option for its full estimate and any measured numbers.