Sub-Net Public
Public benchmarks, tools and releases from Sub-Net e.U.
Sub-Net model benchmarks — reproducible shootouts between self-hosted models (LLMs, VLMs, classifiers).
Updated 2026-09-19 19:31:25 +00:00
Bayesian-optimization autotuner for vLLM serving configs - fork of github.com/SergioMorillas/vllm-bayesian-autotuner with Sub-Net patches (dflash spec-decoding, categorical sampling axes, bare-metal launch, env-key injection hardening)
Updated 2026-09-14 07:56:15 +00:00