// RUN ON CONSUMER GPU

Run Japanese-LM-3.6B 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 → ~773.3 yes 396k ~$1,999
NVIDIA GeForce RTX 4090 → ~435 yes 283k ~$1,599
NVIDIA GeForce RTX 3090 Ti → ~435 yes 283k ~$1,999
NVIDIA GeForce RTX 5080 → ~414.3 yes 171k ~$999
NVIDIA GeForce RTX 3090 → ~403.9 yes 283k ~$1,499
NVIDIA GeForce RTX 3080 Ti → ~393.6 yes 115k ~$1,199
NVIDIA GeForce RTX 5070 Ti → ~386.7 yes 171k ~$749
AMD Radeon RX 7900 XTX → ~366.7 yes 283k ~$999
NVIDIA GeForce RTX 3080 10GB → ~328 yes 87k ~$699
NVIDIA GeForce RTX 4080 → ~317.6 yes 171k ~$1,199
AMD Radeon RX 7900 XT → ~305.6 yes 227k ~$899
NVIDIA GeForce RTX 5070 → ~290 yes 115k ~$549
NVIDIA GeForce RTX 4070 Ti Super → ~290 yes 171k ~$799
NVIDIA GeForce RTX 3070 Ti → ~262.4 yes 59k ~$599
AMD Radeon RX 9070 XT → ~246 yes 171k ~$599
AMD Radeon RX 7800 XT → ~238.3 yes 171k ~$499
NVIDIA GeForce RTX 4070 Super → ~217.5 yes 115k ~$599
NVIDIA GeForce RTX 4070 → ~217.5 yes 115k ~$599
Intel Arc Pro B70 → ~217.1 yes 396k —
Intel Arc A770 16GB → ~200 yes 171k ~$329
AMD Radeon RX 6800 XT → ~195.6 yes 171k ~$649
AMD Radeon RX 6900 XT → ~195.6 yes 171k ~$999
NVIDIA GeForce RTX 3070 → ~193.3 yes 59k ~$499
NVIDIA GeForce RTX 5060 Ti 16GB → ~193.3 yes 171k ~$429
NVIDIA GeForce RTX 5060 Ti 8GB → ~193.3 yes 59k ~$379
Intel Arc Pro B60 → ~162.9 yes 283k —
Intel Arc B580 12GB → ~162.9 yes 115k ~$249
NVIDIA GeForce RTX 4060 Ti 8GB → ~124.3 yes 59k ~$399
NVIDIA GeForce RTX 4060 Ti 16GB → ~124.3 yes 171k ~$499
NVIDIA DGX Spark → ~117.8 yes 1704k ~$3,999
AMD Ryzen AI Max+ 395 → ~97.8 yes 1704k ~$1,999
Intel Arc Pro B50 → ~80 yes 171k —
NVIDIA GeForce RTX 2050 Mobile → ~48.3 tight 3k —

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

→ Japanese-LM-3.6B overview → All hardware

FAQ

What Consumer GPU hardware runs Japanese-LM-3.6B?▶

On GenAIList, 33 Consumer GPU options can run Japanese-LM-3.6B at Q4 — the fastest is the NVIDIA GeForce RTX 5090 at ~773.3 tok/s.

How fast is Japanese-LM-3.6B on Consumer GPU hardware?▶

Estimated generation speed ranges from ~48.3 to ~773.3 tok/s at Q4 (memory-bandwidth bound). Click any option for its full estimate and any measured numbers.