IntelliBar is a macOS menu-bar assistant that puts several AI models in one keyboard-driven interface. It connects directly to model providers using your own accounts or API keys, and to local models through Ollama, so there is no subscription for the app itself and you pay providers for what you use. The homepage lists OpenAI, Anthropic, Google, Grok, Llama, Qwen, DeepSeek and Perplexity Sonar models; that list dates from early 2025, while the public release notes show version 0.47.0, dated 13 January, adding support for newer models including Claude 4.5 Sonnet, GPT-5.2 and Gemini 3 Pro.
The features serve people who use AI dozens of times a day: sending one prompt to several models and comparing the answers side by side, smart completions as you type, a command menu on Command-K, instant search across chat history on Command-O, voice input and read-aloud, and reusable slash commands for common instructions. Privacy is the design pitch: chats are stored locally on the device, prompts go straight to the chosen provider with no intermediary, and local models keep everything on the machine, though anything sent to a cloud model is governed by that provider’s policy.
It is sold as a one-time license, from $39 according to a third-party listing, with a free evaluation period and no usage limits beyond what your providers charge. The developers describe it as user-funded, with no investors, ads or data sales. Being a small independent app, support and roadmap depend on a small team; the regular public release history is the reassuring signal.
