10
Interactive Views
10+
LLM Providers
< 60s
1 GB Parse Time
SQL-like
HeapQL Queries
Why HeapLens?
⚡
High-Performance Rust Engine
Zero-copy mmap parsing with Lengauer-Tarjan dominator tree computation. 1 GB heap dump parsed in ~60 seconds on Apple M1.
🔍
10 Interactive Analysis Views
Overview, Histogram, Dominator Tree, Leak Suspects, Waste Detection, Source Bridging, HeapQL, Compare, Timeline, and AI Chat.
🤖
AI-Native Analysis
Built-in LLM chat supporting 10+ providers including Claude, GPT, Gemini, and local models via Ollama. Ask questions in plain English.
🗄️
HeapQL Query Language
SQL-like queries purpose-built for heap analysis. Filter by class, size, retained size with autocomplete and syntax highlighting.
🔌
MCP Integration
Model Context Protocol server mode lets Claude Desktop and other AI clients analyze heap dumps programmatically.
📊
Compare & Timeline
Diff two heap dumps side-by-side or track memory trends across multiple snapshots with D3.js visualizations.