// RUN ON CONSUMER GPU

Run ProGen2-xlarge 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 → ~435 yes 186k ~$1,999
NVIDIA GeForce RTX 4090 → ~244.7 yes 130k ~$1,599
NVIDIA GeForce RTX 3090 Ti → ~244.7 yes 130k ~$1,999
NVIDIA GeForce RTX 5080 → ~233 yes 74k ~$999
NVIDIA GeForce RTX 3090 → ~227.2 yes 130k ~$1,499
NVIDIA GeForce RTX 3080 Ti → ~221.4 yes 45k ~$1,199
NVIDIA GeForce RTX 5070 Ti → ~217.5 yes 74k ~$749
AMD Radeon RX 7900 XTX → ~206.3 yes 130k ~$999
NVIDIA GeForce RTX 3080 10GB → ~184.5 yes 31k ~$699
NVIDIA GeForce RTX 4080 → ~178.7 yes 74k ~$1,199
AMD Radeon RX 7900 XT → ~171.9 yes 102k ~$899
NVIDIA GeForce RTX 5070 → ~163.1 yes 45k ~$549
NVIDIA GeForce RTX 4070 Ti Super → ~163.1 yes 74k ~$799
NVIDIA GeForce RTX 3070 Ti → ~147.6 yes 17k ~$599
AMD Radeon RX 9070 XT → ~138.4 yes 74k ~$599
AMD Radeon RX 7800 XT → ~134.1 yes 74k ~$499
NVIDIA GeForce RTX 4070 Super → ~122.3 yes 45k ~$599
NVIDIA GeForce RTX 4070 → ~122.3 yes 45k ~$599
Intel Arc Pro B70 → ~122.1 yes 186k —
Intel Arc A770 16GB → ~112.5 yes 74k ~$329
AMD Radeon RX 6800 XT → ~110 yes 74k ~$649
AMD Radeon RX 6900 XT → ~110 yes 74k ~$999
NVIDIA GeForce RTX 3070 → ~108.8 yes 17k ~$499
NVIDIA GeForce RTX 5060 Ti 16GB → ~108.8 yes 74k ~$429
NVIDIA GeForce RTX 5060 Ti 8GB → ~108.8 yes 17k ~$379
Intel Arc Pro B60 → ~91.6 yes 130k —
Intel Arc B580 12GB → ~91.6 yes 45k ~$249
NVIDIA GeForce RTX 4060 Ti 8GB → ~69.9 yes 17k ~$399
NVIDIA GeForce RTX 4060 Ti 16GB → ~69.9 yes 74k ~$499
NVIDIA DGX Spark → ~66.3 yes 840k ~$3,999
AMD Ryzen AI Max+ 395 → ~55 yes 840k ~$1,999
Intel Arc Pro B50 → ~45 yes 74k —

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

→ ProGen2-xlarge overview → All hardware

FAQ

What Consumer GPU hardware runs ProGen2-xlarge?▶

On GenAIList, 32 Consumer GPU options can run ProGen2-xlarge at Q4 — the fastest is the NVIDIA GeForce RTX 5090 at ~435 tok/s.

How fast is ProGen2-xlarge on Consumer GPU hardware?▶

Estimated generation speed ranges from ~45 to ~435 tok/s at Q4 (memory-bandwidth bound). Click any option for its full estimate and any measured numbers.