EcoCompute Should-I-Quantize Estimator for LangChain
io.github.hongping-zh/ecocompute-estimator
Energy impact of LLM quantization, calibrated to measured GPU data. Every answer states its basis.
client:LangChain
transport:streamable-http
tools:4
Install EcoCompute Should-I-Quantize Estimator in LangChain
pip install langchain-mcp-adapters
from langchain_mcp_adapters.client import MultiServerMCPClient
client = MultiServerMCPClient({
"ecocompute-estimator": {
"transport": "streamable_http",
"url": "https://ecocompute-estimator.zhanghongping1982.workers.dev/mcp",
}
})
tools = await client.get_tools()
get_tools() returns the server's tools as LangChain tools, ready for an agent. LangChain docs