Documentation
Everything you need to know about RuneSpoke Hub
AI Provider Configuration
Configure and manage AI providers in RuneSpoke Hub. Support for Claude, GPT-4, Gemini, and more with intelligent routing and cost optimization.
BYOAI (Bring Your Own AI)
RuneSpoke Hub supports multiple AI providers with intelligent routing, cost optimization, and unified API access.
- Use your own API keys and credits
- Intelligent model selection and routing
- Cost tracking and optimization
- Fallback and redundancy support
Why an API key, not “Sign in with OpenAI”?
AI providers (OpenAI, Anthropic, Google, …) authenticate their model APIs with a secret API key, not OAuth. There is no OAuth token that lets a third‑party app call a provider’s model endpoint and bill your account - that’s simply how the providers are built. So “BYOAI” means bring your own API key: you paste it once and RuneSpoke stores it encrypted in the vault and uses it on your behalf.
RuneSpoke’s one‑click OAuth is for dev integrations - GitHub, Linear/Jira, Slack, Google - where the providers do offer OAuth. (Provider sign‑in like “Sign in with ChatGPT” is for identity or vendor CLIs, not for programmatic model calls.) Get a key from your provider’s console (e.g. platform.openai.com/api-keys orconsole.anthropic.com), add billing/credit, then add it in AI FinOps.
Supported AI Providers
Anthropic Claude
Claude 3.5 Sonnet, Claude 3 Opus
Advanced reasoning and code generation with excellent instruction following and safety features.
OpenAI GPT
GPT-4o, GPT-4 Turbo, GPT-3.5
Powerful language model with excellent general knowledge and coding capabilities.
Google Gemini
Gemini Pro, Gemini Ultra
Multimodal AI with strong performance on reasoning and code generation tasks.
Perplexity Sonar
Sonar, Sonar Pro, Sonar Reasoning Pro
Web-grounded answers with live citations. Every response is backed by fresh search results - the only frontier provider that does this natively.
GitHub Copilot
Codex-based code completion
Specialized for code completion and generation with GitHub integration.
Configuration Guide
Step 1: Add API Keys
Configure your AI provider API keys in the RuneSpoke Hub settings.
# Environment Variables ANTHROPIC_API_KEY=sk-ant-... OPENAI_API_KEY=sk-... GOOGLE_AI_API_KEY=AI... GITHUB_TOKEN=ghp_...
Step 2: Model Selection
Configure which models to use for different tasks.
Code Generation
Claude 3.5 Sonnet
Code Review
GPT-4o
Quick Tasks
Gemini Pro
Step 3: Routing Rules
Set up intelligent routing based on task type, cost, and performance.
Advanced Features
Smart Routing
Automatically route requests to the best model based on context, cost, and performance requirements.
Cost Optimization
Track usage and costs across providers. Set budgets and get alerts when limits are reached.
Fallback & Redundancy
Configure fallback providers for high availability. Automatic failover when providers are unavailable.
Analytics & Monitoring
Detailed analytics on model performance, response times, and cost breakdowns across all providers.
Best Practices
Model Selection Tips
- Use Claude 3.5 Sonnet for complex code analysis and reasoning tasks
- Use GPT-4o for general programming questions and explanations
- Use Gemini Pro for fast, cost-effective queries with large context
- Use GitHub Copilot for IDE-integrated code completion
Cost Management
- Set monthly budgets for each provider
- Use cheaper models for simple tasks
- Monitor usage patterns and optimize routing
- Enable caching for repeated queries
Next Steps
Now that you understand AI provider configuration, explore these related topics: