Prompt Hub

Manage, version, and evaluate your AI prompts with Langfuse – your open-source LLM observability platform.

Push your lead data to Langfuse for running evaluations

Create, version, and manage your prompt templates with full history.

Monitor LLM calls, latency, token usage, and debug issues.

Build evaluation datasets with expected outputs for testing.

Run automated evals with LLM-as-a-judge or custom metrics.

Test prompts interactively with different models and parameters.

Configure API keys, team members, and project settings.

Every change to your prompts is tracked. Compare versions, roll back, and A/B test different prompt variations.

Track every LLM call with detailed traces. Monitor costs, latency, and quality across all your AI features.

Langfuse is running locally at http://localhost:3000