modelcostsaver for LangChain
io.github.sachinuppal/modelcostsaver
Offline MCP: predict LLM call cost and pick the cheapest capable model. No keys, no network.
client:LangChain
transport:stdio
runtime:npm
Install modelcostsaver in LangChain
pip install langchain-mcp-adapters
from langchain_mcp_adapters.client import MultiServerMCPClient
client = MultiServerMCPClient({
"modelcostsaver": {
"transport": "stdio",
"command": "npx",
"args": ["-y", "@workswarm/modelcostsaver"],
}
})
tools = await client.get_tools()
get_tools() returns the server's tools as LangChain tools, ready for an agent. LangChain docs