Deep dives into how Reachy Mini works, product news, and stories from the team building it.
7 articles · latest on September 30, 2026
A deep dive into the Reachy Mini stack: one WebRTC connection carrying audio, video and motor control, local or remote, and the signaling, SDK and embedding layers on top of it that let anyone - or their coding agent - build and share robot apps in minutes.
September 30, 2026 · 25 min read
Why new Reachy Mini apps are built with the JS SDK - nothing to install, always up to date, safe by default - and what that means for the Python apps you already know and love.
September 30, 2026 · 5 min read
Reachy Mini's conversation currently runs on a single backend: our own open, OpenAI-Realtime-compatible voice engine. Here's the short version, with a pointer to the full deep dive.
July 15, 2026 · 2 min read
How Reachy Mini streams audio and video. The camera, the microphone array, and GStreamer + WebRTC that let the same code work whether an app runs on the robot, your laptop, or a GPU-backed Hugging Face Space.
June 10, 2026 · 15 min read
The Reachy Mini conversation app can now use tools hosted in public Hugging Face Spaces, called over MCP. Give your robot a new ability - like web search or weather - by adding a Space from the Hub, without editing the app or downloading any code.
June 3, 2026 · 11 min read
Tiny price, small size, huge possibilities. Meet Reachy Mini, an expressive, open-source desktop robot for human-robot interaction, creative coding, and AI experimentation - programmable in Python and priced from $399.
July 9, 2025 · 4 min read
