LinguaAI is a Bring Your Own Key (BYOK) AI assistant that lives in your menubar. Access powerful AI models from any Mac app with a simple keyboard shortcut. Your API keys stay on your device, and your data goes directly to the AI provider—we never see it.
Available for all AI models
Create consistent AI writing tones and styles that are automatically applied to your prompts, ensuring the AI responds in your preferred manner. Configure in Settings → Preferences.
AI remembers conversation history and important information across chats, providing continuity in your interactions. Configure in Settings → Preferences.
Adjust response creativity vs precision. Lower values (0.0-0.4) provide focused, consistent responses. Higher values (0.6-1.0) enable more creative, varied outputs. Configure per shortcut in Shortcut Settings.
Available only for Claude models
Faster responses and cost savings. First message costs +25%, but subsequent messages within 5 minutes are 90% cheaper. Requires minimum 1024 tokens to activate. Automatically enabled for shortcuts (can be disabled per shortcut), enable for chats in AI Providers → Account Management.
Enhanced analysis and problem-solving capabilities. Provides more thoughtful responses but increases processing time and output token costs. Best for complex tasks requiring deep thinking. Enable globally in Settings → Preferences or per shortcut in Shortcut Settings.
Real-time internet information access. Claude can search the web to provide current information and cite sources. Additional costs apply for web search usage. Enable globally in Settings → Preferences or per shortcut in Shortcut Settings.
Features handled automatically
OpenAI GPT-5 includes automatic prompt optimization and caching. No additional configuration needed - the model handles efficiency improvements automatically.
Solution: Check accessibility permissions in System Preferences → Security & Privacy → Accessibility. LinguaAI needs access to monitor keyboard shortcuts.
Solution: Verify your API key is valid and has sufficient credits. Check your internet connection. Some models may have rate limits or temporary outages.
Solution: Enable prompt caching for frequently used shortcuts, use shorter prompts, or consider switching to other AI providers for cost savings. Monitor usage in your Anthropic dashboard.
Solution: Disable reasoning if enabled (it increases processing time). Check your internet connection. Consider using a different AI model. Some complex queries naturally take longer.
Contact us at [email protected] for technical support or questions about LinguaAI.