One prompt to test how "open" an open-source model really is
alex_verem · x · 2026-11-13
The author shares a reusable prompt: with web search enabled, ask an AI to verify any "open source" model's true openness — are the weights downloadable, what does the license restrict, is training code published, and is training data named — citing sources for each answer and saying "not found" where unprovable. His experience: most releases pass the first two checks and fail the rest. He recommends running it before building on any model.
More from Apps
- Dev claims the best voice mode and computer use he's used, built on OpenAI Realtime 2 and Claude — RileyRalmuto · 2026-11-13
- Voice Cursor launches Viki beta, a voice assistant that sees your screen and acts on it — Scobleizer · 2026-11-13
- Solo dev ships logic-puzzle roguelike built with Claude that screenshot-tests itself — TheCodingTutor · 2026-11-13
- Multica launches Custom Themes beta with community theme store — jiayuan_jy · 2026-11-13
- Rippling launches AI scheduling agent that builds shifts and forecasts revenue — andreisavu · 2026-11-13
- Multica launches project management for human + agent teams, turning Claude Code and Codex into teammates — jiayuan_jy · 2026-11-13