llama.app Goes Live: llama.cpp Finally Has a Home
🩺 Summary
llama.cpp is the de-facto standard for running AI locally on CPU/GPU. But it has no official homepage - only GitHub. New users don't know it exists. 113.8K Stars on GitHub, creator @ggerganov.
📝 Details
llama.app is now the official homepage for llama.cpp. Tagline: 'AI that lives on your computer. Open-source, private, always local.' Features:
- Zero API keys, zero telemetry, zero limits
- Pair with Pi coding agent (run `llama serve` then launch Pi)
- Runs on any hardware: laptop to cluster, CPU to GPU
**Install in one line:**
- macOS/Linux: `curl -LsSf https://llama.app/install.sh | sh`
- Windows: `irm https://llama.app/install.ps1 | iex`
- Also: Homebrew / Winget
**Recommended models (out of the box):**
- Qwen3.6-27B — single GPU sweet spot
- Qwen3.6-35B-A3B — 35B MoE, 3B active
- Gemma-4-26B-A4B — Google's desktop MoE
- gpt-oss-20b — OpenAI's open-weight release
💬 Comments (0)