// RUN ON APPLE

Run CogVLM-17B on Apple hardware

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

HardwareEst tok/sFits 8k?Max ctx~Price
Apple M2 Ultra 192GB โ†’ ~68.9 yes 617k ~$6,600
Apple M2 Ultra 64GB โ†’ ~68.9 yes 178k ~$4,000
Apple M1 Ultra 128GB โ†’ ~68.9 yes 397k ~$4,800
Apple M1 Ultra 64GB โ†’ ~68.9 yes 178k ~$3,700
Apple M4 Max 64GB โ†’ ~47 yes 178k ~$3,900
Apple M4 Max 36GB โ†’ ~47 yes 82k ~$3,000
Apple M4 Max 128GB โ†’ ~47 yes 397k ~$4,900
Apple M4 Max 48GB โ†’ ~47 yes 123k ~$3,300
Apple M1 Max 32GB โ†’ ~34.5 yes 68k ~$2,000
Apple M3 Max 48GB โ†’ ~34.5 yes 123k ~$3,200
Apple M3 Max 64GB โ†’ ~34.5 yes 178k ~$3,700
Apple M3 Max 128GB โ†’ ~34.5 yes 397k ~$4,700
Apple M1 Max 64GB โ†’ ~34.5 yes 178k ~$2,600
Apple M2 Max 32GB โ†’ ~34.5 yes 68k ~$2,200
Apple M2 Max 64GB โ†’ ~34.5 yes 178k ~$2,800
Apple M2 Max 96GB โ†’ ~34.5 yes 288k ~$3,300
Apple M3 Max 36GB โ†’ ~25.8 yes 82k ~$2,900

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

FAQ

What Apple hardware runs CogVLM-17B?โ–ถ

On GenAIList, 17 Apple options can run CogVLM-17B at Q4 โ€” the fastest is the Apple M2 Ultra 192GB at ~68.9 tok/s.

How fast is CogVLM-17B on Apple hardware?โ–ถ

Estimated generation speed ranges from ~25.8 to ~68.9 tok/s at Q4 (memory-bandwidth bound). Click any option for its full estimate and any measured numbers.