Magnitude (YC S25) Open-Sources Inference Engine with Up to 2x Faster Decode Than llama.cpp
petrusenko_max · x · 2026-10-01
Magnitude (YC S25) released an open-source inference engine that compiles and tunes kernels on-device, reporting up to 2x faster decode than llama.cpp: +92% on Metal, +19% on CUDA. It runs on Apple Silicon, NVIDIA, AMD, or CPU-only hardware under Apache-2.0.
More from Infra
- Microsoft 365 outage postmortem: how automated health checks amplified the failure — _jaydeepkarale · 2026-11-16
- Commerce Dept set to release remote access rule tonight, reporter says — ShakeelHashim · 2026-11-16
- Liquid AI d1-2B decision model runs on-device: ~68ms gesture reading, 14 decisions/s, zero tokens — AccBalanced · 2026-11-16
- Microsoft Outlook outage silenced chlorine alert emails at German theme park, injuring 15 pool guests — AccBalanced · 2026-11-16
- Ming-Chi Kuo: OpenAI and Anthropic orders extend Google TPU visibility to 2029, ~26M units — BenBajarin · 2026-11-16
- Cerebras core revenue up 70% YoY to $231M, backlog hits $26.1B — AccBalanced · 2026-11-16