Qwen3.8-27B at 144 tok/s on an M5 Max MacBook Pro
英文标题：Qwen3.8-27B at 144 tok/s on an M5 Max MacBook Pro
Meet Inco Splash, open-source inference engine, built around the model and around Apple silicon. Up to 3× the decode speed of Ollama, 2× oMLX, and almost 4× when an agent fans out into sub-agents. Requirements: M3 or newer, macOS 26.4+, 36 GB Get started with a single command: brew install inc
打开原文相关 8 · 机会 4
中文
Meet Inco Splash, open-source inference engine, built around the model and around Apple silicon. Up to 3× the decode speed of Ollama, 2× oMLX, and almost 4× when an agent fans out into sub-agents. Requirements: M3 or newer, macOS 26.4+, 36 GB Get started with a single command: brew install inc
中文全文翻译正在补齐，可先查看下方英文原文。
English / 原文
Meet Inco Splash, open-source inference engine, built around the model and around Apple silicon. Up to 3× the decode speed of Ollama, 2× oMLX, and almost 4× when an agent fans out into sub-agents. Requirements: M3 or newer, macOS 26.4+, 36 GB Get started with a single command: brew install inc