# Open WebUI Integration AVELIN API is fully compatible with [Open WebUI](https://openwebui.com) — the most popular self-hosted AI chat interface with over 60K GitHub stars. Open WebUI gives you a polished ChatGPT-like experience while connecting to both local models and AVELIN's frontier API simultaneously. > **Why AVELIN + Open WebUI?** Mix local privacy with frontier intelligence — run small models locally via Ollama for sensitive tasks, then switch to AVELIN's 1M-token context models for complex work, all from the same beautiful interface. Open WebUI (formerly Ollama WebUI) is a self-hosted, feature-rich AI chat platform that supports both local models through Ollama and remote OpenAI-compatible APIs at the same time. With over 60K GitHub stars, it's the most widely adopted self-hosted chat interface in the AI ecosystem. It includes built-in RAG with document upload and retrieval, web search integration, a model playground for comparing outputs, pipeline support for chaining operations, and function calling — all wrapped in a clean, responsive UI that rivals commercial offerings. The killer feature is dual-mode operation: local and cloud simultaneously. Run Llama 3 on your own hardware for private conversations, then switch to `avelin-coding-pro` when you need to analyze an entire codebase with 1M-token context. Your data stays local when it matters, and you tap frontier capability when you need it. It's for developers, researchers, teams, and privacy-conscious users who want a self-hosted ChatGPT-quality experience without sacrificing access to the most capable models available. --- ## Quick Setup ### Step 1: Get your AVELIN API key Sign up at [avelin.ai](https://avelin.ai) and generate an API key. It will look like `sk-ave...xxxx`. ### Step 2: Install Open WebUI ```bash # Docker (recommended) docker run -d -p 3000:8080 \ --add-host=host.docker.internal:host-gateway \ -v open-webui:/app/backend/data \ --name open-webui \ --restart always \ ghcr.io/open-webui/open-webui:main ``` Open WebUI runs at `http://localhost:3000` by default. ### Step 3: Configure AVELIN as a Provider 1. Log in and open the **Admin Panel** (shield icon) 2. Go to **Settings** > **Connections** 3. Under **OpenAI API**, click **Add** 4. Enter your credentials: | Field | Value | |---|---| | **URL** | `https://api.avelin.ai/v1` | | **API Key** | `sk-ave...xxxx` | 5. Click **Save** ### Step 4: Start Chatting 1. Open the model selector in the chat interface 2. Choose an AVELIN model (e.g., `avelin-pro`) 3. Start chatting — or mix with your local Ollama models --- ## Recommended Models ### For General Chat: `avelin-pro` | Feature | avelin-pro | Why it matters for Open WebUI | |---|---|---| | **Context window** | **256K tokens** | Long conversations with full context retention | | **Max output** | **65K tokens** | Detailed, thorough responses | | **Speed** | **Fast** | Responsive streaming in the WebUI | ### For Code & RAG: `avelin-coding-pro` | Feature | avelin-coding-pro | Why it matters for Open WebUI | |---|---|---| | **Context window** | **1M tokens** | Ingest entire repos via RAG document upload | | **Max output** | **65K tokens** | Generate complete files and implementations | | **Reasoning** | Deep step-by-step | Reliable code generation and analysis | ### Quick Reference | Use case | Model | |---|---| | Daily chat & questions | `avelin-pro` | | Complex reasoning | `avelin-ultra` | | Code generation & review | `avelin-coding-pro` | | Architecture & design | `avelin-coding-ultra` | | Fast responses | `avelin-fast` | | Agent/tool workflows | `avelin-agentic-ultra` | | Quick tool calls | `avelin-agentic-fast` | --- ## Why AVELIN for Open WebUI? - **Local + cloud hybrid**: Keep Ollama for private tasks, add AVELIN for frontier capability - **Prompt caching**: ~80% cost reduction on repeated system prompts and RAG contexts - **Automatic failover**: AVELIN switches providers if one is slow or down — your WebUI stays responsive - **Model diversity**: 9 specialized models through one API key, all available in the model selector - **Function calling**: AVELIN's agentic models support Open WebUI's tools and pipelines --- ## Advanced Configuration ### Pipelines Open WebUI supports pipelines for chaining operations. Use AVELIN's agentic models for complex multi-step workflows: 1. Go to **Admin Panel** > **Settings** > **Pipelines** 2. Configure a pipeline with `avelin-agentic-ultra` as the reasoning model 3. Chain with local models for cost optimization on simpler steps ### Web Search Enable web search to give AVELIN models real-time context: 1. Go to **Admin Panel** > **Settings** > **Web Search** 2. Configure your search provider (Google, Bing, DuckDuckGo, etc.) 3. AVELIN models will use retrieved results as context automatically ### Model Comparison (Playground) Use Open WebUI's playground to compare AVELIN models side by side: 1. Open the **Playground** from the sidebar 2. Add `avelin-pro` and `avelin-ultra` as two separate models 3. Send the same prompt to both and compare outputs --- ## Troubleshooting | Issue | Solution | |---|---| | **No AVELIN models showing** | Verify URL is `https://api.avelin.ai/v1` in Connections settings | | **"Invalid API key"** | Check key starts with `sk-ave` and has no extra whitespace | | **Slow responses** | Switch to `avelin-fast` for quicker replies | | **Streaming not working** | Ensure streaming is enabled in Admin Panel > Settings > General | | **RAG not finding documents** | Check document was fully indexed; try `avelin-coding-pro` for 1M context | | **Function calling fails** | Use `avelin-agentic-ultra` which has strong tool-use support | | **Connection timeout** | Check Docker networking; ensure container can reach external APIs | --- ## Related - [API Quickstart](../../api/quickstart.md) — First API call in 5 minutes - [LibreChat Guide](librechat.md) — Another self-hosted chat alternative - [LobeChat Guide](lobechat.md) — Modern self-hosted option with plugins - [Model Catalog](../../models/README.md) — Full model comparison - [Open WebUI Docs](https://docs.openwebui.com) — Official documentation