Integrations
To explore production traces and metrics in Langfuse, you need to integrate your application with Langfuse.
Objective:
- Capture traces of your application
- Add scores to these traces to measure/evaluate quality of outputs
There are currently four main ways to integrate with Langfuse:
| Integration | Supports | Description |
|---|---|---|
| Langchain | JS/TS, Python | Automated instrumentation by passing callback handler to Langchain application. |
| SDK | JS/TS, Python | Manual instrumentation using the SDKs for full flexibility. |
| OpenAI | Python | Automated instrumentation using drop-in replacement of OpenAI SDK. |
| API | Directly call the public API. OpenAPI spec available. |
In addition, Langfuse is natively integrated with the following projects/packages:
| Name | Description |
|---|---|
| Flowise | JS/TS no-code builder for customized LLM flows. |
| Langflow | Python-based UI for LangChain, designed with react-flow to provide an effortless way to experiment and prototype flows. |
| LiteLLM | Use any LLM as a drop in replacement for GPT. Use Azure, OpenAI, Cohere, Anthropic, Ollama, VLLM, Sagemaker, HuggingFace, Replicate (100+ LLMs). |
Unsure which integration to choose? Ask us on Discord or in the chat.