alt Hacker NewsWhy it needs to work only on a Mac? And why is that better than running the gpt oss with llama.cpp and codex on my Linux box?
Why it needs to work only on a Mac? And why is that better than running the gpt oss with llama.cpp and codex on my Linux box?
Our model is bigger and more capable than gpt OSS and can run at full context at 40 tokens / s.
We are rolling out to Mac to start with plans to release windows and Linux within 3 months.