Claude Code + NEO: Ship production AI from your terminal

VS CodeCursor<br>Sign InGet Started

Extend Claude limits<br>by offloading AI tasks to Neo

1Install neo-mcp

Add NEO's MCP server to any environment with Python 3.11+.<br>Terminal<br>pip install neo-mcp

2Create your Secret Key

Open your NEO dashboard, create a key, and copy it. Keys look like sk-v1-….<br>Get Access Key

3Connect Claude Code

Register NEO with one command, then just ask in a new chat to ship work.<br>Command<br>claude mcp add --scope user neo \<br>-e NEO_SECRET_KEY=sk-v1-your-key \<br>-- python3 -m neo_mcp

Using Cursor, VS Code, or another MCP client? See the neo-mcp setup.

Ship production AI directly

claude session<br>Your browser doesn't support embedded video. Download the demo.

Independent benchmark<br>62% cheaper. Better answer.<br>The task: benchmark a speech-to-text model on a CPU-only Azure VM — 2 cores, 7.7 GB RAM, no GPU. Claude Code alone reached for the obvious HuggingFace + PyTorch path and iterated in real time. NEO spent two minutes researching first, then chose ONNX Runtime for its AVX2-optimized CPU kernels — same task, same machine.

Cost per task<br>$1.96→$0.74<br>62% cheaper

Runtime (RTF)<br>0.519→0.328<br>37% faster

Backend chosen<br>PyTorch→ONNX Runtime<br>CPU-optimized

Benchmark by Gaurav Vij · Read the full writeup

Start shipping.<br>Install once, then delegate ML work to NEO from any Claude Code session.<br>pip install neo-mcpRead the setup guide