PS5LM: llama.cpp port turns a jailbroken PS5 into a home LLM server (500+ GitHub stars)
wateriscoding · x · 2026-10-18
A developer open-sourced PS5LM, which runs llama.cpp natively on a jailbroken PS5's GPU, turning the console into a private home model server. The repo already has 500+ stars.
- Runs any GGUF model (starting with Qwen 3 3.8B) and exposes OpenAI- and Anthropic-compatible APIs so every device on the network can use it
- Ships a TV-friendly UI aimed at jailbroken-PS5 owners who want a local model host
- Newest release adds Ollaya-powered decision models (d1, Clef, OpenJev) via the same /v1/systemone API, returning typed decisions with calibrated probabilities
More from Fun
- GPU Furnace: Thermostat-Controlled GPU Workloads That Heat Your Room — PraxisOG · 2026-10-18
- New site detects if your website was vibe coded by reading its design fingerprints — sahilypatel · 2026-10-18
- Astra "grammar corrections" turn out to be fixing pronouns in a paper — Sauers_ · 2026-10-18
- AI art feud escalates: anti-AI artist RJ Palmer called out for selling Pokemon merch — AIandDesign · 2026-10-18
- Creator builds 700+ page world bible with ChatGPT over 4 years, turns it into an image-gen "taste" profile — thederbiedone · 2026-10-18
- Grok matter-of-factly lists which groups user was racist against, exempting Africans and Latin Americans — teortaxesTex · 2026-10-18