// RUN ON CONSUMER GPU

Run MiniCPM-1.2B 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 → ~2,231.8 yes 416k ~$1,999
NVIDIA GeForce RTX 4090 → ~1,255.4 yes 303k ~$1,599
NVIDIA GeForce RTX 3090 Ti → ~1,255.4 yes 303k ~$1,999
NVIDIA GeForce RTX 5080 → ~1,195.6 yes 191k ~$999
NVIDIA GeForce RTX 3090 → ~1,165.7 yes 303k ~$1,499
NVIDIA GeForce RTX 3080 Ti → ~1,135.8 yes 135k ~$1,199
NVIDIA GeForce RTX 5070 Ti → ~1,115.9 yes 191k ~$749
AMD Radeon RX 7900 XTX → ~1,058.2 yes 303k ~$999
NVIDIA GeForce RTX 3080 10GB → ~946.5 yes 107k ~$699
NVIDIA GeForce RTX 4080 → ~916.6 yes 191k ~$1,199
AMD Radeon RX 7900 XT → ~881.8 yes 247k ~$899
NVIDIA GeForce RTX 5070 → ~836.9 yes 135k ~$549
NVIDIA GeForce RTX 4070 Ti Super → ~836.9 yes 191k ~$799
NVIDIA GeForce RTX 3070 Ti → ~757.2 yes 79k ~$599
AMD Radeon RX 9070 XT → ~709.9 yes 191k ~$599
AMD Radeon RX 7800 XT → ~687.8 yes 191k ~$499
NVIDIA GeForce RTX 4070 Super → ~627.7 yes 135k ~$599
NVIDIA GeForce RTX 4070 → ~627.7 yes 135k ~$599
Intel Arc Pro B70 → ~626.7 yes 416k —
Intel Arc A770 16GB → ~577.2 yes 191k ~$329
AMD Radeon RX 6800 XT → ~564.4 yes 191k ~$649
AMD Radeon RX 6900 XT → ~564.4 yes 191k ~$999
NVIDIA GeForce RTX 3070 → ~557.9 yes 79k ~$499
NVIDIA GeForce RTX 5060 Ti 16GB → ~557.9 yes 191k ~$429
NVIDIA GeForce RTX 5060 Ti 8GB → ~557.9 yes 79k ~$379
Intel Arc Pro B60 → ~470 yes 303k —
Intel Arc B580 12GB → ~470 yes 135k ~$249
NVIDIA GeForce RTX 4060 Ti 8GB → ~358.7 yes 79k ~$399
NVIDIA GeForce RTX 4060 Ti 16GB → ~358.7 yes 191k ~$499
NVIDIA DGX Spark → ~340 yes 1724k ~$3,999
AMD Ryzen AI Max+ 395 → ~282.2 yes 1724k ~$1,999
Intel Arc Pro B50 → ~230.9 yes 191k —
NVIDIA GeForce RTX 2050 Mobile → ~139.5 yes 23k —

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

→ MiniCPM-1.2B overview → All hardware

FAQ

What Consumer GPU hardware runs MiniCPM-1.2B?▶

On GenAIList, 33 Consumer GPU options can run MiniCPM-1.2B at Q4 — the fastest is the NVIDIA GeForce RTX 5090 at ~2,231.8 tok/s.

How fast is MiniCPM-1.2B on Consumer GPU hardware?▶

Estimated generation speed ranges from ~139.5 to ~2,231.8 tok/s at Q4 (memory-bandwidth bound). Click any option for its full estimate and any measured numbers.