AI_Eli 0.1.1

AI_Eli, the earliest version of ELI: a local voice assistant for Linux that runs on Ollama.

Get started

1. Install Ollama and pull any chat model:

   curl -fsSL https://ollama.com/install.sh | sh
   ollama pull qwen2.5:1.5b

2. Download AI_Eli-0.1.1-x86_64.AppImage, make it executable and open it:

   chmod +x AI_Eli-0.1.1-x86_64.AppImage
   ./AI_Eli-0.1.1-x86_64.AppImage

Everything else is inside the AppImage: Python and all its libraries, the desktop windows, the voice assistant and the Whisper speech model, so speech recognition works offline. AI_Eli talks to the newest chat model in your Ollama unless you name another.

Command
./AI_Eli-0.1.1-x86_64.AppImage	the desktop window
./AI_Eli-0.1.1-x86_64.AppImage --voice	the hands-free voice assistant (say "Eli" first)
./AI_Eli-0.1.1-x86_64.AppImage --documents	the PDF and memory window
./AI_Eli-0.1.1-x86_64.AppImage --terminal	chat in the terminal

Its own folders and settings

AI_Eli keeps everything in its own folders, apart from any other ELI version: what it remembers in ~/.local/share/AI_Eli, settings in ~/.config/AI_Eli/config as KEY=VALUE lines (the README lists them), for example ELI_MODEL=llama3.2. A variable named with AI_ELI_ in place of ELI_ works too: AI_ELI_MODEL=llama3.2 ./AI_Eli-0.1.1-x86_64.AppImage. Plain ELI_ variables belong to other ELI versions and are ignored.

Fixed since 0.1.0

• 0.1.0 kept its settings and chat state in ~/.eli_config, ~/.eli_state.json and ~/.eli_memory.jsonl, names other ELI versions could share. They are in AI_Eli's own folders now.
• The window saved conversations to whatever folder it was started from; they go to AI_Eli's folder now.
• Settings exported for another ELI version (such as ELI_FULL_CONTROL) no longer reach AI_Eli.
• The speech model's onnxruntime queued usage telemetry for Microsoft. That is switched off.

Needs

• Linux on x86_64 with glibc 2.35 or newer (Ubuntu 22.04, Debian 12, Fedora 36 or later)
• PulseAudio or PipeWire, and espeak-ng for the spoken replies
• If your system will not open AppImages, run it with --appimage-extract-and-run

SHA256SUMS.txt holds the file's checksum. Apache 2.0.