llm-analytics-setup

You instrument LLM analytics by wrapping your LLM calls with PostHog tracking using the provided provider references. This setup captures $ai_generation events and traces across platforms like OpenAI and Anthropic directly in your codebase.

Tracking LLM generations across multiple providers is best handled by implementing standardized $ai_generation events. This approach uses provider-specific references to ensure consistent analytics capture across different platforms.

Yes, PostHog LLM analytics instrumentation works with frameworks like LangChain, LlamaIndex, and Vercel AI. Provider reference files include setup guidance for wrapping calls within these specific frameworks.

You can monitor local LLM calls with Ollama using this instrumentation. The provided references include setup documentation for Ollama alongside other providers to capture generation analytics.

You verify LLM analytics instrumentation by checking for $ai_generation traces and generations within your PostHog dashboard. The setup guidance includes specific verification steps to confirm events are logging correctly.