InferSSD on a 256 GB Mac Studio, Tested: Plain llama.cpp Wrote Three Times Faster
0 View
Share this Video
Publish Date:
2 November, 2026
Category:
CNBC
Video License
Standard License
Imported From:
Youtube
Tags
By Evan Vega
Novel Cognition ran WiCi’s open-source InferSSD engine and the standard llama.cpp engine on the same 473 GiB DeepSeek V4.1 Flash model on a Mac Studio with 256 GB of memory. llama.cpp wrote answers at 10.95 tokens a second against InferSSD’s best of 3.45; InferSSD read prompts as fast or faster, and its 200 GB setting never ran. WiCi’s published 48 GB laptop result is a different machine and is untouched.
