BlockRun
Get started
Integrations / LangChain

Use BlockRun as a LangChain chat model through a local sidecar

Run the blockrun-litellm sidecar and point LangChain's ChatOpenAI at it: tools, streaming and async all work, and every call pays per request in USDC over x402.

Get your agent on BlockRun Read the docs

How it works

BlockRun's chat endpoint is already OpenAI-compatible; the only difference from api.openai.com is authentication — a per-request wallet signature instead of a bearer key. The blockrun-litellm sidecar bridges exactly that: a local OpenAI-compatible proxy that signs x402 payments with your wallet, so ChatOpenAI works unchanged, including tool calling, streaming and async.

Prefer no sidecar? An in-process path registers BlockRun as a LiteLLM provider and uses ChatLiteLLM, and a minimal custom LLM class over the Python SDK covers text-in, text-out chains and RAG. The sidecar is the recommended path because it keeps the full chat-model feature set.

Behind it is the BlockRun catalog: 78 chat models plus image, video, music and speech generation, search, market data and multi-chain RPC, priced per call. 6 open-weight models are free and need no wallet.

Setup

1. Install

pip install 'blockrun-litellm[proxy]' langchain langchain-openai

2. Start the sidecar

export BLOCKRUN_WALLET_KEY=0x<base wallet key>
blockrun-litellm-proxy --port 4001

3. Point ChatOpenAI at it

from langchain_openai import ChatOpenAI
llm = ChatOpenAI(model="openai/gpt-5.4", base_url="http://127.0.0.1:4001/v1", api_key="dummy")
print(llm.invoke("Explain x402 in one sentence").content)

Questions

Can I pay on Solana instead of Base?
Yes — start the sidecar with the Solana API URL and a SOLANA_WALLET_KEY.
Is the sidecar exposed to the network?
It binds to loopback by default; set BLOCKRUN_PROXY_TOKEN before binding anywhere else.

Related