The QuadB64 family is like a superhero team of encodings, each with a unique power to solve specific data problems while always preventing accidental matches. Whether you need to perfectly preserve data, find similar items, or map locations, there’s a QuadB64 hero for the job.
QuadB64 Encoding Family
The QuadB64 family consists of multiple encoding schemes, each optimized for specific use cases:
Core Encodings
Q64 - Quadrant Base64
The foundational encoding that uses position-dependent alphabets to ensure substring safety.
EQ64 - Extended Quadrant Base64
Enhanced version with optimized alphabet rotation for better distribution.
ShQ64 - Sharded Quadrant Base64
Designed for distributed systems with built-in sharding support.
T8Q64 - Top-8 Bits Quadrant Base64
Optimized for sparse high-dimensional vectors, prioritizing the most significant bits.
ZoQ64 - Zoned Quadrant Base64
Supports variable-length embeddings with zone-based encoding.
MQ64 - Matryoshka Quadrant Base64
Implements nested resolution layers for progressive refinement.
Choosing an Encoding Method
| Method | Best For | Key Feature |
|---|---|---|
| Q64 | General use | Balanced performance |
| EQ64 | High-throughput systems | Optimized alphabets |
| ShQ64 | Distributed databases | Built-in sharding |
| T8Q64 | Sparse vectors | Bit prioritization |
| ZoQ64 | Variable embeddings | Flexible zones |
| MQ64 | Progressive search | Multi-resolution |
Common Characteristics
All QuadB64 encodings share these properties:
- Position-safe: No false substring matches
- Search-friendly: Optimized for text indexing
- Reversible: Lossless encoding/decoding
- Compact: Efficient space utilization
- Fast: SIMD-optimized implementations