LangDB
LangDB is an AI gateway and governance platform that routes requests across 250+ models from providers such as OpenAI, Anthropic, Google, Meta, Mistral, and DeepSeek through a single project-scoped, OpenAI-compatible REST API. It layers routing, guardrails, tracing, cost control, and an MCP (Model Context Protocol) gateway on top of that unified interface.
APIs
LangDB Chat Completions / Routing API
OpenAI-compatible chat completions routed across 250+ models from OpenAI, Anthropic, Google, Meta, Mistral, and DeepSeek, with SSE streaming, tool calling, structured outputs, f...
LangDB Embeddings API
OpenAI-compatible embeddings endpoint that creates vector representations of input text or token arrays via any embedding model routed through the gateway.
LangDB Images API
OpenAI-compatible image generation endpoint producing images from text prompts using image-capable models routed through the gateway.
LangDB Models API
Lists the models available to a project through the gateway in an OpenAI-compatible shape, spanning 250+ models across the supported providers.
LangDB Threads / Messages API
Retrieves the ordered messages and aggregated per-thread cost for a conversation thread, grouped by the X-Thread-Id supplied on chat requests.
LangDB Analytics / Usage API
Returns time-series analytics, aggregated summaries, total usage, and per-model usage (cost and token counts) for a project over preset reporting windows.
LangDB MCP Gateway API
Attaches virtual and custom MCP servers (over SSE or WebSocket transport) to chat requests via the mcp_servers array, exposing their tools to the routed model and bundling multi...
Event Specifications
LangDB Chat Completions Streaming (HTTP + SSE)
AsyncAPI 2.6 description of LangDB's **chat completion streaming** surface. LangDB does not publish a WebSocket API for chat. The asynchronous / event-style transport documented...
ASYNCAPI