Implementation:Langfuse Langfuse API LLM Connections Schema

Knowledge Sources	Langfuse Langfuse API Reference
Domains	API, LLM Connections
Last Updated	2026-02-14 00:00 GMT

Overview

Fern API definition for the LLM connections API, enabling projects to configure and manage API connections to various LLM providers such as OpenAI, Anthropic, Azure, Bedrock, and Google AI.

Description

This file defines the public API contract for managing LLM connections under /api/public. LLM connections store the configuration needed to interface with different LLM providers for features like playground evaluation. The file contains two endpoints:

• GET /api/public/llm-connections -- List all LLM connections in a project (paginated, secrets excluded)
• PUT /api/public/llm-connections -- Create or update an LLM connection (upserts on provider name)

The definition includes an LlmAdapter enum supporting six provider adapters: Anthropic, OpenAI, Azure, Bedrock, Google Vertex AI, and Google AI Studio. Adapter-specific configuration is supported via the config field (required for Bedrock, optional for Vertex AI).

Usage

Developers reference this definition when:

• Adding support for new LLM providers
• Generating SDK types for LLM connection management
• Implementing the playground or eval features that use LLM connections
• Understanding how secret keys and extra headers are handled in the API

Code Reference

Source Location

• Repository: Langfuse
• File: fern/apis/server/definition/llm-connections.yml
• Lines: 1-112

Signature

service:
  auth: true
  base-path: /api/public
  endpoints:
    list:
      method: GET
      docs: Get all LLM connections in a project
      path: /llm-connections
      response: PaginatedLlmConnections

    upsert:
      method: PUT
      docs: Create or update an LLM connection. The connection is upserted on provider.
      path: /llm-connections
      request: UpsertLlmConnectionRequest
      response: LlmConnection

types:
  LlmAdapter:
    enum:
      - value: anthropic
        name: Anthropic
      - value: openai
        name: OpenAI
      - value: azure
        name: Azure
      - value: bedrock
        name: Bedrock
      - value: google-vertex-ai
        name: GoogleVertexAI
      - value: google-ai-studio
        name: GoogleAIStudio

I/O Contract

Endpoints

Method	Path	Description
GET	/api/public/llm-connections	List all LLM connections in a project (paginated, secrets excluded)
PUT	/api/public/llm-connections	Create or update an LLM connection (upserts on provider name)

Key Types

Type Name	Description
LlmConnection	Connection record with id, provider, adapter, masked secret key, baseURL, custom models, extra header keys, config, timestamps
PaginatedLlmConnections	Paginated list of LlmConnection with MetaResponse
UpsertLlmConnectionRequest	Request with provider, adapter, secretKey, baseURL, customModels, withDefaultModels, extraHeaders, config
LlmAdapter	Enum: anthropic, openai, azure, bedrock, google-vertex-ai, google-ai-studio

Usage Examples

# List LLM connections
curl -X GET "https://cloud.langfuse.com/api/public/llm-connections?page=1&limit=10" \
  -H "Authorization: Bearer $API_KEY"

# Create/update an OpenAI connection
curl -X PUT "https://cloud.langfuse.com/api/public/llm-connections" \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "provider": "openai",
    "adapter": "openai",
    "secretKey": "sk-...",
    "customModels": ["gpt-4o-mini"],
    "withDefaultModels": true
  }'

# Create a Bedrock connection (config required)
curl -X PUT "https://cloud.langfuse.com/api/public/llm-connections" \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "provider": "bedrock",
    "adapter": "bedrock",
    "secretKey": "aws-secret-key",
    "config": {"region": "us-east-1"}
  }'

Related Pages

• Environment:Langfuse_Langfuse_Node_24_Runtime