langchain-nuroute

A LangChain chat model for NuRoute — a provider-agnostic LLM gateway that routes, monitors, and controls inference requests across OpenAI, Anthropic, Google, and more.

This talks to the gateway through the official aicp client rather than through ChatOpenAI's OpenAI-compatibility path, so NuRoute-specific behavior — the "auto" routing model, provider overrides, and NuRouteError — comes through unchanged instead of being flattened into OpenAI's response/error shapes.

Installation

pip install langchain-nuroute

Requires Python 3.9+.

Quick start

from langchain_nuroute import ChatNuRoute
from langchain_core.messages import HumanMessage

llm = ChatNuRoute(
    nuroute_api_key="aicp-...",
    nuroute_base_url="https://your-nuroute-gateway",
    model="auto",  # classifies the request and routes it — see nuroute.ai/docs for strategies
)

response = llm.invoke([HumanMessage(content="Hello!")])
print(response.content)

Streaming

for chunk in llm.stream([HumanMessage(content="Tell me a story")]):
    print(chunk.content, end="", flush=True)

Forcing a provider or model

llm = ChatNuRoute(
    nuroute_api_key="aicp-...",
    nuroute_base_url="https://your-nuroute-gateway",
    model="claude-sonnet-4-6",
    provider="anthropic",
    temperature=0.2,
    max_tokens=1024,
)

Usage and routing metadata

ChatResult.llm_output carries the model NuRoute actually routed to and the token usage for the call:

result = llm.generate([[HumanMessage(content="Hello!")]])
print(result.llm_output["model"], result.llm_output["usage"])

Documentation

Full SDK documentation: nuroute.ai/docs

License

MIT