// RUN ON CONSUMER GPU

Run GPT-NeoX-20B on Consumer GPU hardware

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

HardwareEst tok/sFits 8k?Max ctx~Price
NVIDIA GeForce RTX 5090 → ~139.2 yes 64k ~$1,999
NVIDIA GeForce RTX 4090 → ~78.3 yes 36k ~$1,599
NVIDIA GeForce RTX 3090 Ti → ~78.3 yes 36k ~$1,999
NVIDIA GeForce RTX 5080 → ~74.6 tight 8k ~$999
NVIDIA GeForce RTX 3090 → ~72.7 yes 36k ~$1,499
NVIDIA GeForce RTX 5070 Ti → ~69.6 tight 8k ~$749
AMD Radeon RX 7900 XTX → ~66 yes 36k ~$999
NVIDIA GeForce RTX 4080 → ~57.2 tight 8k ~$1,199
AMD Radeon RX 7900 XT → ~55 yes 22k ~$899
NVIDIA GeForce RTX 4070 Ti Super → ~52.2 tight 8k ~$799
AMD Radeon RX 9070 XT → ~44.3 tight 8k ~$599
AMD Radeon RX 7800 XT → ~42.9 tight 8k ~$499
Intel Arc Pro B70 → ~39.1 yes 64k —
Intel Arc A770 16GB → ~36 tight 8k ~$329
AMD Radeon RX 6800 XT → ~35.2 tight 8k ~$649
AMD Radeon RX 6900 XT → ~35.2 tight 8k ~$999
NVIDIA GeForce RTX 5060 Ti 16GB → ~34.8 tight 8k ~$429
Intel Arc Pro B60 → ~29.3 yes 36k —
NVIDIA GeForce RTX 4060 Ti 16GB → ~22.4 tight 8k ~$499
NVIDIA DGX Spark → ~21.2 yes 391k ~$3,999
AMD Ryzen AI Max+ 395 → ~17.6 yes 391k ~$1,999
Intel Arc Pro B50 → ~14.4 tight 8k —

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

→ GPT-NeoX-20B overview → All hardware

FAQ

What Consumer GPU hardware runs GPT-NeoX-20B?▶

On GenAIList, 22 Consumer GPU options can run GPT-NeoX-20B at Q4 — the fastest is the NVIDIA GeForce RTX 5090 at ~139.2 tok/s.

How fast is GPT-NeoX-20B on Consumer GPU hardware?▶

Estimated generation speed ranges from ~14.4 to ~139.2 tok/s at Q4 (memory-bandwidth bound). Click any option for its full estimate and any measured numbers.