Dev builds all-rust AI OS LexiSuite, claims faster Qwen3 27B inference than Ubuntu
W61k3r · reddit · 2026-11-12
A Reddit user released LexiSuite, an all-rust AI operating system bundling an inference engine, custom kernel, and tuning/management/chat/training panel, shipped as a headless bootable USB image that loads models from Hugging Face.
The author claims Qwen3 27B runs faster on a 7900 XTX than llama under the same config on Ubuntu, and the stack supports llama.cpp and vLLM; a T8 test image with a debug kit is out for testers. Caveat: the source is in a private repo as the author weighs funding options, so the performance claims aren't independently verifiable yet.
More from Infra
- Agent Substrate Calculator Estimates Session Resume Speed Under Load — pjausovec · 2026-11-12
- Dell expands AI Data Platform: agents have a data-context problem, not a model problem — DavidLinthicum · 2026-11-12
- SGLang Megakernel Pushes Muse Spark to 1,109 tok/s, 2.1x Over Baseline — hsu_byron · 2026-11-12
- Potion UI's local login system turns your PC into a multi-user AI server — cocktailpeanut · 2026-11-12
- AWS Finally Built Its Own AWS Emulator — super_linear · 2026-11-12
- Megakernel is multiprogramming for GPUs: 'GPUs are entering the CPU age' — pranjalssh · 2026-11-12