PS5LM runs llama.cpp on a jailbroken PS5, hitting ~20 tok/s with Qwen 3.8 27B
kimmonismus · x · 2026-10-14
PS5LM runs llama.cpp on a jailbroken PlayStation 5, first on its Zen 2 CPU and then GPU. Target: Qwen 3.8 27B (hybrid Gated DeltaNet) at 20 tok/s. v0.1 ships a payload (ps5lm.elf) opening a model library in the PS5 browser to download, run and chat; currently small models up to 1.6GB (Qwen3.5 0.8B/2B), with the 27B on the roadmap.
More from Fun
- Nobel laureate Henri Kagan, 95, honored for a breakthrough he made at 55 — PeterDiamandis · 2026-10-15
- Dev's parody take on getting blocked: 'too hot and handsome and smart' — threepointone · 2026-10-15
- Agent breaks prod, then fixes it itself: 502, Coolify logs, self-recovery — tristanbob · 2026-10-15
- With zero motion design background, dev crafts Apple-style launch film in 3 days using Opus 5.5 — Glittering-Art3723 · 2026-10-15
- "Now That We Know They All Distilled, It's Even Funnier": AI Circle Reacts to China Model Claims — scaling01 · 2026-10-15
- Dev builds AI-vs-AI RTS game using Codex for art and MiniMax for sprite animations — emersoftware · 2026-10-15