# Vane | Kushagra Srivastava | Knowledge Base | Instagit

Vane is an AI-powered answering engine.

GitHub Stars: 32.7k

Repository: https://github.com/ItzCrazyKns/Vane

---

## Articles

### [What Are Vane's API Routes Used For? A Complete Guide to the Backend Architecture](/ItzCrazyKns/Vane/what-are-vanes-api-routes-used-for)

Explore Vane's API routes to understand its backend architecture. Learn how it validates requests loads LLM providers executes agents and streams data to your frontend.

- Tags: architecture
- Published: 2026-03-11

### [How Vane Uses Embedding Models for Semantic Search: Architecture and Implementation](/ItzCrazyKns/Vane/how-does-vane-use-embedding-models-for-semantic-search)

Discover how Vane leverages embedding models for semantic search. Learn about its architecture and implementation for efficient document and query vectorization and ranking.

- Tags: architecture
- Published: 2026-03-11

### [How Vane Provides Fully Cited Answers: A Technical Architecture Breakdown](/ItzCrazyKns/Vane/what-enables-vane-to-provide-cited-answers)

Discover Vane's technical architecture. Learn how XML tags and strict prompts enable Vane to generate fully cited answers and enforce citation requirements for factual statements.

- Tags: architecture
- Published: 2026-03-11

### [How to Connect Vane to an External SearxNG Instance: Complete Configuration Guide](/ItzCrazyKns/Vane/can-i-connect-vane-to-an-external-searxng-instance)

Easily connect Vane to an external SearxNG instance using the SEARXNG_API_URL environment variable or config.json. Configure your SearxNG URL and enhance your search experience now.

- Tags: how-to-guide
- Published: 2026-03-11

### [How SearxNG Is Bundled in Vane's Docker Deployment: Complete Technical Architecture](/ItzCrazyKns/Vane/how-is-searxng-bundled-in-vanes-docker-deployment)

Discover how SearxNG is bundled in Vane's Docker deployment. Learn about the technical architecture, virtual environment setup, and parallel startup process.

- Tags: architecture
- Published: 2026-03-11

### [Where is the SearxNG Configuration File Located in Vane? A Complete Guide](/ItzCrazyKns/Vane/where-is-the-searxng-configuration-file-located-in-vane)

Find the SearxNG configuration file location in Vane. Discover its path in the repository and within the Docker container for easy access and modification.

- Tags: how-to-guide
- Published: 2026-03-11

### [How Vane Integrates with SearxNG: Architecture and Implementation Guide](/ItzCrazyKns/Vane/how-does-vane-integrate-with-searxng)

Learn how Vane integrates with SearxNG using its three-layer architecture. Discover environment-aware config, Dockerization, and a TypeScript HTTP wrapper for standardized search queries.

- Tags: architecture
- Published: 2026-03-11

### [What Search Backend Does Vane Use? SearXNG Integration Explained](/ItzCrazyKns/Vane/what-search-backend-does-vane-use)

Discover Vane's search backend integration with SearXNG. Learn how Vane fetches web results through a flexible HTTP endpoint for enhanced search experiences.

- Tags: internals
- Published: 2026-03-11

### [How to Configure Vane's LLM Providers and Models: A Complete Guide](/ItzCrazyKns/Vane/how-can-i-configure-vanes-llm-providers-and-models)

Learn how to configure Vane's LLM providers and models with this complete guide. Master Vane's three-layer architecture for seamless integration and model loading.

- Tags: how-to-guide
- Published: 2026-03-11

### [Where to Find the Vane Model Registry Implementation: Complete Guide](/ItzCrazyKns/Vane/where-can-i-find-the-vane-model-registry-implementation)

Locate the Vane model registry implementation in the ModelRegistry class at ItzCrazyKns/Vane. Discover provider loading, model discovery, and lifecycle management features.

- Tags: how-to-guide
- Published: 2026-03-11

### [How Vane Integrates with Ollama for Local LLMs: A Technical Deep Dive](/ItzCrazyKns/Vane/how-does-vane-integrate-with-ollama-for-local-llms)

Discover how Vane integrates with Ollama for local LLMs. This technical deep dive explains the provider architecture, model discovery, and official npm client usage for seamless local AI.

- Tags: deep-dive
- Published: 2026-03-11

### [Which LLM Providers Are Currently Supported by Vane?](/ItzCrazyKns/Vane/which-llm-providers-are-currently-supported-by-vane)

Discover which LLM providers Vane supports including OpenAI, Ollama, Google Gemini, and more. Integrate seamlessly with popular models in your projects.

- Tags: faq
- Published: 2026-03-11

### [Understanding the Base Interface for LLM and Embedding Models in Vane](/ItzCrazyKns/Vane/what-is-the-base-interface-for-llm-and-embedding-models-in-vane)

Discover the Base Interface for LLM and Embedding Models in Vane. Learn about abstract base classes and provider implementation requirements.

- Tags: internals
- Published: 2026-03-11

### [How Vane Defines Classification Prompts for LLM Search Routing](/ItzCrazyKns/Vane/how-are-classification-prompts-defined-for-vanes-llm)

Learn how Vane defines classification prompts for LLM search routing. Discover the `classifierPrompt` system prompt and its role in web search, personal document queries, and more.

- Tags: how-to-guide
- Published: 2026-03-11

### [Where Is the Vane Question Classification Logic Implemented?](/ItzCrazyKns/Vane/where-is-the-vane-question-classification-logic-implemented)

Discover where Vane's question classification logic resides in the search agent at src/lib/agents/search/classifier.ts. Learn how LLM prompts categorize queries for efficient search operations.

- Tags: internals
- Published: 2026-03-11

### [Vane Question Processing Workflow: From API Request to Streamed Answer](/ItzCrazyKns/Vane/what-is-the-high-level-workflow-for-processing-a-question-in-vane)

Understand the Vane question processing workflow from API request to streamed answers. Explore its seven-stage pipeline for intent classification, parallel widgets, iterative research, and token-level responses.

- Tags: architecture
- Published: 2026-03-11

### [How Vane's Agent Orchestration Works: A Deep Dive into the Search Pipeline](/ItzCrazyKns/Vane/how-does-vanes-agent-orchestration-work)

Explore Vane's agent orchestration pipeline. Learn how it classifies intent, runs parallel widgets, and uses an iterative LLM research loop for streamed answers. Discover the Vane search flow.

- Tags: deep-dive
- Published: 2026-03-11

### [What Are the Seven Key Components of Vane's Architecture?](/ItzCrazyKns/Vane/what-are-the-seven-key-components-of-vanes-architecture)

Explore Vane's architecture. Discover the seven essential components including UI, API, Agents, Search, LLMs, Embedding Models, and Storage. Learn how these TypeScript modules power Vane's functionality.

- Tags: architecture
- Published: 2026-03-11

### [Where to Find Vane's Architecture Documentation: Complete Developer Guide](/ItzCrazyKns/Vane/where-can-i-find-vanes-architecture-documentation)

Find Vane's architecture documentation in the ItzCrazyKns/Vane repository docs directory. Access READMEmd for overviews and WORKINGmd for detailed request flows. Your complete developer guide.

- Tags: architecture
- Published: 2026-03-11

### [How Vane Combines LLMs and Web Search: Inside the Researcher Agent Architecture](/ItzCrazyKns/Vane/how-does-vane-combine-llms-and-web-search)

Discover how Vane integrates LLMs and web search. Its Researcher agent iteratively uses SearxNG for tool-calling, returning cited results for enhanced reasoning.

- Tags: architecture
- Published: 2026-03-11

### [Core Architecture of the Vane AI Answering Engine: A Technical Deep Dive](/ItzCrazyKns/Vane/what-is-the-core-architecture-of-the-vane-ai-answering-engine)

Explore the core architecture of the Vane AI answering engine. Learn about its modular, streaming-first design, React frontend, Node.js backend, and real-time response capabilities.

- Tags: deep-dive
- Published: 2026-03-11

### [How Vane Ensures User Privacy with AI Answering: A Technical Architecture Breakdown](/ItzCrazyKns/Vane/how-does-vane-ensure-user-privacy-with-ai-answering)

Discover how Vane ensures user privacy with AI answering using local databases, environment variables, offline LLMs, and self-hosted search. Learn the technical architecture.

- Tags: architecture
- Published: 2026-03-11