// RUN ON CONSUMER GPU

Run base LM+GNN+kNN on Consumer GPU hardware

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

HardwareEst tok/sFits 8k?Max ctx~Price
NVIDIA GeForce RTX 5090 โ†’ ~10,160.6 yes 424k ~$1,999
NVIDIA GeForce RTX 4090 โ†’ ~5,715.3 yes 312k ~$1,599
NVIDIA GeForce RTX 3090 Ti โ†’ ~5,715.3 yes 312k ~$1,999
NVIDIA GeForce RTX 5080 โ†’ ~5,443.2 yes 199k ~$999
NVIDIA GeForce RTX 3090 โ†’ ~5,307.1 yes 312k ~$1,499
NVIDIA GeForce RTX 3080 Ti โ†’ ~5,171 yes 143k ~$1,199
NVIDIA GeForce RTX 5070 Ti โ†’ ~5,080.3 yes 199k ~$749
AMD Radeon RX 7900 XTX โ†’ ~4,817.5 yes 312k ~$999
NVIDIA GeForce RTX 3080 10GB โ†’ ~4,309.2 yes 115k ~$699
NVIDIA GeForce RTX 4080 โ†’ ~4,173.1 yes 199k ~$1,199
AMD Radeon RX 7900 XT โ†’ ~4,014.6 yes 256k ~$899
NVIDIA GeForce RTX 5070 โ†’ ~3,810.2 yes 143k ~$549
NVIDIA GeForce RTX 4070 Ti Super โ†’ ~3,810.2 yes 199k ~$799
NVIDIA GeForce RTX 3070 Ti โ†’ ~3,447.3 yes 87k ~$599
AMD Radeon RX 9070 XT โ†’ ~3,231.8 yes 199k ~$599
AMD Radeon RX 7800 XT โ†’ ~3,131.4 yes 199k ~$499
NVIDIA GeForce RTX 4070 Super โ†’ ~2,857.7 yes 143k ~$599
NVIDIA GeForce RTX 4070 โ†’ ~2,857.7 yes 143k ~$599
Intel Arc Pro B70 โ†’ ~2,853 yes 424k โ€”
Intel Arc A770 16GB โ†’ ~2,627.7 yes 199k ~$329
AMD Radeon RX 6800 XT โ†’ ~2,569.3 yes 199k ~$649
AMD Radeon RX 6900 XT โ†’ ~2,569.3 yes 199k ~$999
NVIDIA GeForce RTX 3070 โ†’ ~2,540.1 yes 87k ~$499
NVIDIA GeForce RTX 5060 Ti 16GB โ†’ ~2,540.1 yes 199k ~$429
NVIDIA GeForce RTX 5060 Ti 8GB โ†’ ~2,540.1 yes 87k ~$379
Intel Arc Pro B60 โ†’ ~2,139.7 yes 312k โ€”
Intel Arc B580 12GB โ†’ ~2,139.7 yes 143k ~$249
NVIDIA GeForce RTX 4060 Ti 8GB โ†’ ~1,633 yes 87k ~$399
NVIDIA GeForce RTX 4060 Ti 16GB โ†’ ~1,633 yes 199k ~$499
NVIDIA DGX Spark โ†’ ~1,547.9 yes 1733k ~$3,999
AMD Ryzen AI Max+ 395 โ†’ ~1,284.7 yes 1733k ~$1,999
Intel Arc Pro B50 โ†’ ~1,051.1 yes 199k โ€”
NVIDIA GeForce RTX 2050 Mobile โ†’ ~635 yes 31k โ€”

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

FAQ

What Consumer GPU hardware runs base LM+GNN+kNN?โ–ถ

On GenAIList, 33 Consumer GPU options can run base LM+GNN+kNN at Q4 โ€” the fastest is the NVIDIA GeForce RTX 5090 at ~10,160.6 tok/s.

How fast is base LM+GNN+kNN on Consumer GPU hardware?โ–ถ

Estimated generation speed ranges from ~635 to ~10,160.6 tok/s at Q4 (memory-bandwidth bound). Click any option for its full estimate and any measured numbers.