Faster
- The Agents and Providers pages open at once again. With many providers, the Providers page could take 25 seconds and the Agents page 7: each agent's model and routing group was worked out again for every provider, large config files were parsed over and over (Codex's
config.tomlcan hold thousands of trusted projects), and Alma was asked for its providers several times per look. Now the Providers page loads in about 0.02s and the Agents page in about 0.1s. - A config file is parsed again only when it changes. What Alma is on is reused for a few seconds when only showing it; anything magpie changes in Alma still reads Alma fresh first.
Install
Download from usemagpie.ai, or curl -fsSL https://usemagpie.ai/install.sh | sh. An installed magpie updates itself.
- macOS (Apple Silicon):
magpie-darwin-arm64.dmg - macOS (Intel):
magpie-darwin-amd64.dmg - Windows:
magpie-windows-amd64.exe(ARM:magpie-windows-arm64.exe) - Linux:
magpie-linux-amd64/magpie-linux-arm64(needs GTK 3 and WebKitGTK 4.1) - Terminal only:
magpie-cli-<os>-<arch>