// RUN ON CONSUMER GPU

Run Tri-21B 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 → ~134.3 yes 62k ~$1,999
NVIDIA GeForce RTX 4090 → ~75.5 yes 34k ~$1,599
NVIDIA GeForce RTX 3090 Ti → ~75.5 yes 34k ~$1,999
NVIDIA GeForce RTX 5080 → ~71.9 tight 6k ~$999
NVIDIA GeForce RTX 3090 → ~70.1 yes 34k ~$1,499
NVIDIA GeForce RTX 5070 Ti → ~67.1 tight 6k ~$749
AMD Radeon RX 7900 XTX → ~63.7 yes 34k ~$999
NVIDIA GeForce RTX 4080 → ~55.2 tight 6k ~$1,199
AMD Radeon RX 7900 XT → ~53.1 yes 20k ~$899
NVIDIA GeForce RTX 4070 Ti Super → ~50.4 tight 6k ~$799
AMD Radeon RX 9070 XT → ~42.7 tight 6k ~$599
AMD Radeon RX 7800 XT → ~41.4 tight 6k ~$499
Intel Arc Pro B70 → ~37.7 yes 62k —
Intel Arc A770 16GB → ~34.7 tight 6k ~$329
AMD Radeon RX 6800 XT → ~34 tight 6k ~$649
AMD Radeon RX 6900 XT → ~34 tight 6k ~$999
NVIDIA GeForce RTX 5060 Ti 16GB → ~33.6 tight 6k ~$429
Intel Arc Pro B60 → ~28.3 yes 34k —
NVIDIA GeForce RTX 4060 Ti 16GB → ~21.6 tight 6k ~$499
NVIDIA DGX Spark → ~20.5 yes 389k ~$3,999
AMD Ryzen AI Max+ 395 → ~17 yes 389k ~$1,999
Intel Arc Pro B50 → ~13.9 tight 6k —

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

→ Tri-21B overview → All hardware

FAQ

What Consumer GPU hardware runs Tri-21B?▶

On GenAIList, 22 Consumer GPU options can run Tri-21B at Q4 — the fastest is the NVIDIA GeForce RTX 5090 at ~134.3 tok/s.

How fast is Tri-21B on Consumer GPU hardware?▶

Estimated generation speed ranges from ~13.9 to ~134.3 tok/s at Q4 (memory-bandwidth bound). Click any option for its full estimate and any measured numbers.