Rackside

Rackside — 27 Aug

• Ollama v0.33.1: what changed for a home GPU

  why it matters: a local runtime bump you can apply without a cloud account

  source: github.com/ollama/ollama/releases

• llama.cpp releases still ship CPU/GPU builds for self-hosters

  why it matters: the fallback stack when a vendor runtime lags

  source: github.com/ggml-org/llama.cpp/releases

Why today: two local-runtime pages moved; nothing here needs an API key

Watch later: @rackside_brief_bot