Liquid-Codec decoder ported to Apple Neural Engine, 6x faster at 316x real time
JosephJacks_ · x · 2026-11-14
A developer converted Liquid AI's Liquid-Codec speech decoder (24 kHz, 12 codebooks) to Core ML to run on the Apple Neural Engine. On a MacBook Air M3 with int8 quantization it decodes at 316x real time vs 53x on CPU — roughly 6x faster. The project is open-sourced as liquidcodec-ane on GitHub with conversion scripts, examples, and models, under the LFM Open License v1.0 (commercial use limited to companies under $10M annual revenue).
More from Infra
- Developer now runs 100% of his agents locally on a Mac Mini, up from 25% earlier this year — josh_bickett · 2026-11-14
- Nvidia's $530B backstop for AI customers: the six assumptions that must hold — iamKierraD · 2026-11-14
- RTX 5070 Ti ComfyUI benchmarks: overclocking and launch flags yield surprising results — AnalElder · 2026-11-14
- Bernstein: Neoclouds earn ~6x more revenue per megawatt than Bitcoin miners — Beth_Kindig · 2026-11-14
- ZONOS3-Mini: MoE TTS model runs realtime on a Pixel 9 and MacBook Air — cocktailpeanut · 2026-11-14
- The Periodic Table of Inference: 12 LLM serving concepts as meme elements — sophiamyang · 2026-11-14