v0.6.11 — 60× faster regex, path-boosted ranking

Semantic code search that
never phones home.

Ask questions in English. Get answers in code.
100% local, lightning fast, zero telemetry.

Terminal — ig

# Ask in English, find the code intuitively
$ ig "where is tax calculated?"
✓ Found 3 semantic matches in 45ms
src/finance/billing.rs:182
    pub fn calculate_sales_tax(amount: f64, rate: f64) -> f64 {
src/checkout/tax.rs:44
    fn apply_vat_rate(subtotal: Decimal, region: &Region) -> Decimal {
src/invoice/total.rs:301
    let tax_total = subtotal * self.region.tax_rate;

Capabilities

Why developers choose ivygrep

The speed of grep with the understanding of an IDE — all running locally.

⚡

Blazing Fast

Sub-100ms queries on repos with millions of lines. Hash embeddings deliver instant indexing; neural models upgrade silently in the background.

🔒

100% Local Privacy

No API keys. No cloud backend. No telemetry. Your proprietary codebase never leaves your machine — ever.

🧠

Natural Language

Combines classic BM25 text matching with 384-dimensional quantized ONNX neural embeddings. Ask "where is the tax calculated?" and get the right function.

🌿

Git-Native

Understands branches, worktrees, and .gitignore natively. Branch switches trigger instant Merkle-diff re-indexes — never a full re-scan.

🤖

Built for AI Agents

Ships with a built-in MCP server. One command to connect Claude Code, Cursor, Codex, Gemini, or OpenCode — no plugins needed.

🏗️

Code-Aware Chunking

20 languages with Tree-sitter AST parsing. Understands functions, classes, and imports — not just lines of text.

How ivygrep compares

Feature	`grep` / `rg`	GitHub Search	`zoekt`	ivygrep
Works offline	✅	❌	✅	✅
Natural language queries	❌	⚠️	❌	✅
Semantic understanding	❌	❌	❌	✅
Sub-100ms latency	✅	❌	✅	✅
Privacy-first (no upload)	✅	❌	✅	✅
AST code chunking	❌	❌	⚠️	✅
Incremental Merkle indexing	❌	❌	❌	✅
Git worktree overlays	❌	❌	❌	✅
Built-in MCP server	❌	❌	❌	✅

Under the Hood

Two-Tier Hybrid Architecture

Fast indexing with progressive quality upgrades — you never wait for neural models.

Hash Index

Sub-second indexing via 128-bit xxh3 SIMD structural hashes. Instant search availability.

Hybrid Fusion

Lexical BM25 (Tantivy) and semantic ANN (USearch) fuse via Reciprocal Rank Fusion for high-precision results.

Neural Enhancement

Background daemon silently upgrades to quantized ONNX (AllMiniLML6V2) with Apple CoreML / CUDA acceleration.

Git-Native Intelligence

Deeply understands git. Not an afterthought — a core design decision.

Worktree Overlays

When you git worktree add, ivygrep does not copy the index. It creates a thin overlay storing only divergent chunks & tombstones — referencing the base index. A 92K-file repo with 50 changed files creates a ~200 KB overlay instead of duplicating ~50 MB.

Branch-Switch Deltas

Switching branches triggers a targeted Merkle-diff re-index of only the changed files — no full re-scan. Content-based hashes mean byte-identical files across branches are never re-processed.

Benchmarks

Speed That Scales

Benchmarked on the Linux kernel (92K files, 1.5M chunks) and 2GB+ monorepos (289K files, 3.8M chunks)

grep -rnexact string

~9.0 s

1×

rgexact string

~2.7 s

3×

igsemantic search

~72 ms

125×

ig --literalsingle identifier

~17 ms

529×

ig --regex2GB+ monorepo

~200 ms

60×

grep and rg scan every file at query time. ivygrep queries a pre-built index — searches are orders of magnitude faster on warm repos, and semantic mode finds related code even when you don't know the exact identifier.

× faster than grep

~17ms

Single identifier lookup

~72ms

Semantic: "kernel memory allocation"

2.3×

Index-to-source ratio

Concurrent Search (AI Agent Load)

8 threads × Linux kernel — simulating MCP agent traffic

Linux Ryzen 9 (Avg)

11 ms

Mac M4 Max (Avg)

26 ms

Linux Ryzen 9 (p95)

15 ms

Mac M4 Max (p95)

62 ms

Language Support

44 Languages, 20 Tree-sitter ASTs

Tree-sitter languages get function/class-level structural chunking. All others get intelligent heuristic chunking with full index coverage.

🌲 Tree-sitter AST 20 languages

Rust Python Go JavaScript TypeScript Java C C++ C# Scala PHP Ruby Swift Bash Haskell OCaml Lua Dart Objective-C Perl

📐 Heuristic Chunking 24 more

Kotlin Zig Nim Groovy Elixir Erlang Clojure R Julia PowerShell SQL GraphQL HTML CSS Protobuf Terraform TOML YAML JSON Dockerfile Makefile + unknown exts

Model Context Protocol

Supercharge Your AI Agents

ivygrep ships a built-in MCP server. Instead of stuffing entire repos into context windows, your agent queries your indexed codebase natively — with sub-100ms latency.

Claude Code

claude mcp add -s user ig -- ig --mcp

Codex

codex mcp add ig -- ig --mcp

Gemini CLI

gemini mcp add --transport stdio ig ig --mcp

Cursor

"ig": { "command": "ig", "args": ["--mcp"] }

OpenCode

opencode mcp add

How MCP Search Works

🤖

Agent sends query

"find error handling logic"

→

🔍

ig searches index

BM25 + Vector + Literal

→

📄

Returns top chunks

Function bodies, classes, imports

→

✅

Agent gets context

Without loading entire files

Get Started

Install ivygrep

Two commands to semantic search bliss.

🍺 Homebrew (macOS / Linux)

$ brew tap bvolpato/tap
$ brew install bvolpato/tap/ivygrep

📦 Pre-built Binaries

# Download from GitHub Releases
$ curl -L https://github.com/bvolpato/ivygrep/releases/latest/download/ig-$(uname -m)-$(uname -s | tr A-Z a-z).tar.gz | tar xz
$ sudo mv ig /usr/local/bin/

Quick Start

Index your project

ig --add .

Search with natural language

ig "where is authentication handled?"

Or find exact identifiers

ig --literal "handleAuth"

Connect AI agents

claude mcp add -s user ig -- ig --mcp