---

🧩HilbertSFC

Ultra-fast 2D & 3D Hilbert space-filling curve encode/decode kernels for Python.

_{2D Hilbert curves (nbits 1..5) and 3D Hilbert curves (nbits 1..4, animated).}

---

This project is performance-first and implemented entirely in Python. The hot kernels are JIT-compiled with Numba and tuned for:

• Branchless, fully unrolled inner loops
• SIMD via LLVM vector intrinsics
• Small, L1-cache-friendly lookup tables (LUTs)
• Reduced dependency chains for better ILP, e.g., state-independent gather
• Optional multi-threading for batch operations

It provides both convenient Python APIs and kernel accessors designed to be embedded into other Numba kernels.

Quick start

Start here: Quick start

API reference

• hilbert_encode_2d
• hilbert_decode_2d
• hilbert_encode_3d
• hilbert_decode_3d

For kernel accessors, cache helpers, and typing aliases, use the navigation bar.