by Langfuse
Langfuse is an open-source AI engineering platform for debugging, analyzing and improving AI agent applications. It covers tracing, prompt management, evaluation and an AI gateway.
Web appMCP serverFreemium
Step 1: Create a Langfuse Cloud account or self-host.
Step 2: Create API keys in the project settings.
Step 3: Install an SDK or the OpenAI drop-in wrapper and set the keys as environment variables.
Step 4: Run your app so traces appear in Langfuse.
Step 5: Add prompts, scores and dashboards as needed.
Yes. The docs say it is open source and self-hostable, and the instrumentation code is the same for Cloud and self-hosted.
The docs list EU, US, Japan and HIPAA regions, each with its own base URL.
Through native Python and JS SDKs, 100+ framework integrations, OpenTelemetry, or an LLM gateway such as LiteLLM.
It does not block threats itself. Run-time libraries such as LLM Guard handle that, while Langfuse traces and scores help you check how well those measures work.
Official Langfuse Agent Skill for tracing, prompt management, evaluation, datasets, scores, API workflows, and documentation.
DB-GPTDB-GPT is an AI-focused web app for data workflows, giving users practical model-powered capabilities through a dedicated product or developer interface.
Keywords AIRespan is an open-source AI gateway with built-in observability and evals. It routes requests to 1,000+ models through one API and traces, monitors and tests every agent run.