LangChain managed models - Docs by LangChain

Private beta: The LLM Gateway is in private beta. APIs and features may change as we iterate. Sign up for the waitlist to get access.

The /langchain provider serves LangChain-managed models through the LLM Gateway. With the other providers, you bring your own account and key. With managed models, LangChain owns the upstream provider credential, so you can start calling a model immediately with only your LangSmith API key. No provider secret setup is required.

Base URL

Point any OpenAI-compatible client at:

https://gateway.smith.langchain.com/langchain/v1

Authenticate with your LangSmith API key as a bearer token.For regional base URLs, see Regional gateways.

Available models

The provider is OpenAI-compatible and advertises branded model IDs. You select a model by name (case-insensitive) in the request body, and the gateway routes it to the managed upstream model.

Model ID	Description
`LangSmith:pro`	A high-capability general-purpose managed model.

You can also list the available models programmatically:

curl https://gateway.smith.langchain.com/langchain/v1/models \
    -H "Authorization: Bearer $LANGSMITH_API_KEY"

The response is the standard OpenAI /v1/models list containing only the managed model IDs above.

Prerequisites

Before you can call the /langchain provider, confirm that:

Your organization is on a paid plan (Developer, Plus, Startup, or Premier).
You have a workspace-scoped LangSmith API key attached to a role with gateway:invoke and workspaces:read permissions. See Admin setup if you’re unsure.

Make a call

Point any OpenAI-compatible client at https://gateway.smith.langchain.com/langchain/v1, authenticate with your LangSmith API key, and set model to one of the advertised model IDs.

curl https://gateway.smith.langchain.com/langchain/v1/chat/completions \
    -H "Authorization: Bearer $LANGSMITH_API_KEY" \
    -H "Content-Type: application/json" \
    -d '{"model":"LangSmith:pro","messages":[{"role":"user","content":"ping"}]}'

import os

from openai import OpenAI

client = OpenAI(
    base_url="https://gateway.smith.langchain.com/langchain/v1",
    api_key=os.environ["LANGSMITH_API_KEY"],
)
response = client.chat.completions.create(
    model="LangSmith:pro",
    messages=[{"role": "user", "content": "ping"}],
)
print(response.choices[0].message.content)

import os

from langchain.chat_models import init_chat_model

model = init_chat_model(
    model="LangSmith:pro",
    model_provider="openai",
    base_url="https://gateway.smith.langchain.com/langchain/v1",
    api_key=os.environ["LANGSMITH_API_KEY"],
)
print(model.invoke("ping").content)

Supported endpoints

The /langchain provider forwards two OpenAI-compatible prompt endpoints and serves the model list locally:

Method & path	Behavior
`POST /langchain/v1/chat/completions`	Chat completions, including streaming.
`POST /langchain/v1/responses`	OpenAI Responses API.
`GET /langchain/v1/models`	Lists the managed model IDs.

Any other path returns 404. A request for a model that isn’t advertised also returns 404 with a message listing the available model IDs.

Pricing

Invocations will be billed to your account. Standard gateway spend policies still apply on top of the managed-model cap, so you can add tighter per-key or per-user limits.

Tracing

Like all gateway traffic, managed-model calls are traced to LangSmith. See Traces, Engine, and access control for where traces land and how to control access to them.

Next steps

Quickstart: make your first gateway-proxied call.
Spend policies: add cost limits on top of the managed-model cap.
Custom model providers: route to your own OpenAI-compatible endpoints.

Connect these docs to Claude, VSCode, and more via MCP for real-time answers.

Edit this page on GitHub or file an issue.

Base URL

​Available models

​Prerequisites

​Make a call

​Supported endpoints

​Pricing

​Tracing

​Next steps

Available models

Prerequisites

Make a call

Supported endpoints

Pricing

Tracing

Next steps