Version 2.1 Documentation Center
Master System-Wide
Privacy-First AI Refinement
Explore step-by-step installation guides, provider integration manuals, global hotkey shortcuts, and troubleshooting steps for Avelyn.
Documentation Categories
Select a category below to explore topics, configuration steps, and tutorials.
Core Features
- Prompt EnhancementAutomatically refine vague queries into high-precision prompts.
- Grammar CorrectionInstant rule-based and neural grammar and syntax fixing.
- Professional WritingElevate tone, clarity, and phrasing for executive communication.
- Email WritingDraft crisp, persuasive emails directly inside Mail or Outlook.
- AI RewriteParaphrase, rephrase, expand, or condense highlighted text.
- TranslationTranslate between 50+ languages system-wide without tab-switching.
- SummarizationCondense long articles, PDFs, and meeting notes into key bullet points.
- Coding AssistantRefactor, explain, and debug code snippets inside any editor.
AI Providers
- Gemini IntegrationConnecting Google Gemini Flash for ultra-fast cloud responses.
- Ollama IntegrationRunning local LLMs (Gemma 3, Llama 3, Qwen) completely offline.
- OpenRouter IntegrationAccessing Claude 3.5, GPT-4o, DeepSeek V3 via OpenRouter API key.
- Custom API SetupConfiguring LM Studio, vLLM, Groq, or OpenAI-compatible endpoints.
- Smart RouterTask-based automatic routing to specialized models.
- Auto ProviderDynamic provider selection based on query length and privacy rules.
Settings & Configuration
- General SettingsLaunch on startup, tray behavior, and auto-replace toggles.
- Hotkey ConfigurationCustomizing global hotkeys and push-to-talk triggers.
- Model ConfigurationSetting default models, temperature, and system prompts.
- Context WindowManaging token budgets (2k / 4k / 8k context limits).
- Theme & AppearanceDark, Light, and Glassmorphic aesthetic customization.
Advanced Customization
- Custom System PromptsCreating custom mode presets tailored to your domain.
- Context Length OptimizationToken budgeting strategies for low-memory Macs.
- Smart Router ConfigFine-tuning task group mappings (Coding, Writing, Reasoning).
- Auto Provider ConfigAdjusting character thresholds and privacy fallbacks.
- Smart FallbackConfiguring multi-provider priority failovers.
Troubleshooting
- Ollama Connection IssuesResolving server port 11434 and daemon startup errors.
- API Key ProblemsFixing 401 unauthorized, rate-limiting, and key validation issues.
- Performance OptimizationEliminating latency spikes and warm GPU idle routines.
- Permission IssuesResetting macOS Accessibility and Input Monitoring permissions.
- Model Loading ErrorsHandling missing weights and VRAM allocation failures.
Changelog & Releases
Popular Quick Tutorials