alt Hacker NewsSpeaking of llama and local compute, there was a tweet from Georgi Gerganov (llama.cpp author) a couple of days ago saying that he is currently using Qwen3.6 27B, running locally on a Mac M2 Ultra or RTX 5090, to assist with llama.cpp development.
Speaking of llama and local compute, there was a tweet from Georgi Gerganov (llama.cpp author) a couple of days ago saying that he is currently using Qwen3.6 27B, running locally on a Mac M2 Ultra or RTX 5090, to assist with llama.cpp development.