Splash 1.1.0 released, GGUF quants support, MLX import and more
retained text
On my M5 Pro 64GB I can comfortably work in an agentic setup with the Qwen3.8 27B model in good quality (Unsloth UD-Q4\_K\_XL) at a decent speed of 50 t/s. Splash combines optimized kernels, excellent speculative decoding, a well-implemented prefix cache, and mixed-weight support in a single program. To me, this is a breakthrough in local inference on Apple Silicon. [https://github.com/incoai/splash/releases/tag/1.1.0](https://github.com/incoai/splash/releases/tag/1.1.0)
document fetched 2026-09-27 00:39 · canonical https://reddit.com/r/LocalLLaMA/comments/1wqw9rn/splash_110_released_gguf_quants_support_mlx