llama.cpp launches llama.app: frontier AI fully local with one command
ngxson · x · 2026-09-13
llama.cpp (127.8K GitHub stars) launched its official site llama.app, positioning AI as fully local: no API keys, no telemetry, your models and data stay on your machine.
- One-line curl installer, plus Brew/Winget/source builds
- llama serve + pi-llama plugin lets the local coding agent Pi auto-discover your model with zero config; files never leave the machine
- Same binary and hand-tuned kernels from laptops (Apple Silicon) to clusters (H100, MI300, B200, DGX Spark)
- Site features runnable models including Qwen 3.8, DeepSeek V4, Gemma 4, GLM 4.7, and Poolside Laguna XS 2.1
More from coding & agent
- New survey maps recursive self-improvement in agents by autonomy stages — dair_ai · 2026-09-13
- New Cursor CLI sandbox escape variant runs arbitrary code on your Mac — EdenEmarco177 · 2026-09-13
- Garry Tan: Most New YC Apps Are Domain-Specific AI Agents Built on Systems of Record — garrytan · 2026-09-13
- Agent builds 15x14km 3D tile of Marbella in ~1h: 4,393 houses, 2,492 pools, 450km drivable roads — LinusEkenstam · 2026-09-13
- Indie dev explores building a game entirely with the Astra agent, sees 10k-agent studios ahead — LinusEkenstam · 2026-09-13
- Building a dune racer with the Astra agent: per-wheel suspension, drifting physics, Unreal port planned — LinusEkenstam · 2026-09-13