GoKawiilmere.run · Docs · Downloads · Relay + Nodes · Plugins · swift build
mere.run is a local-first inference runtime for Apple Silicon and headless Linux. One public CLI covers image, text, speech, vision, music, sound, video, 3D, worlds, training, and local API serving. The optional macOS Studio uses that same CLI, model store, and run history rather than a separate backend.
One real terminal session on an M4 Max — chat, image, grounding, image→3D, SFX, music, speech round-trip, video, API serving, and a typed workflow graph. Everything local.
Start here
Install a signed release from mere.run/downloads, then let the CLI inspect the machine before pulling a large model:
mere.run setup mere.run model capabilities --recommended mere.run guide --list
For a small first image workflow:
mere.run model pull image-zimage-nano mere.run image generate \ --model image-zimage-nano \ --prompt " a ceramic mug in soft morning light " \ --output ./mug.png
Use mere.run guide <command path> for the packaged offline cookbook behind a creative or model-specific command, and mere.run <group> --help for the exact current flags.
What works today
... continue reading