YieldOS-Lite – A simulator for LLM inference control-plane governance
Simulates governance policies without CUDA kernels or real vLLM schedulers.
Train the smallest LM you can that fits in 16MB. Best model wins!
24M params in 15MB using GPTQ-lite and Muon optimizer for OpenAI's Parameter Golf challenge.
ML researchers, model optimization engineers, constraint optimization enthusiasts
NanoGPT Speedrun · Parameter Golf Challenge
Simulates governance policies without CUDA kernels or real vLLM schedulers.
Yet another markdown editor competing with iA Writer, Bear, and Obsidian.
15MB YouTube music player with native audio threading beats Electron bloat entirely.
Free UX survey tool, but Typeform and specialized UX platforms already do this.
24M spending rows searchable instantly; missing data but fills a real transparency gap.
258 parameters controlling 2.2M connections with 8,384:1 compression is wild.