One MCP. Plug into Cursor, Claude Code, or any agent. Faster edits, smarter retrieval, and better context.
Powered by three specialized models
Lightning-fast code merging
Apply AI-generated edits at 10,500+ tokens per second. No waiting, no lag—just instant code updates that keep your flow uninterrupted.
20x faster grep
Sub-agent built for grepping that finds the right code without polluting context. Precision retrieval at lightning speed.
Semantic code understanding
Embeddings trained on millions of commits. Outperforms OpenAI, Voyage, and Qwen on real world code retrieval benchmarks.