llmfit v1.1.10 Adds RamaLama Runtime Discovery and Publishes the First MLX vs llama.cpp Metal Head-to-Head on Identical Hardware

GitHub·medium signalXBlueskyLinkedInCopy link

AlexsJones/llmfit (32,008 stars, Rust) released v1.1.10 on 2026-08-17, adding RamaLama runtime discovery to its MCP server, the Qwen3.8 model family, and MiniMax M3 vision capability exposure. The release also merged 32 MLX benchmark results on an Apple M4 Pro — the project's first MLX entries — giving a head-to-head against llama.cpp Metal on identical hardware, alongside fixes so one sized Ollama install no longer marks an entire model family installed and MLX community tags resolve to catalog IDs. For anyone choosing a local runtime on Apple Silicon, this is the rare apples-to-apples number.