Randomized trial demonstrates high-performance quantum error correction decoding, suggesting advantages for specific codes.
qector-decoder-v3 (current PyPI release v0.6.6) is a source-available Rust-backed Python platform for quantum error-correction (QEC) decoding. It provides a uniform API over exact minimum-weight perfect matching (BlossomDecoder), fast approximate Union-Find and its SIMD-accelerated variant (FastUnionFindDecoder), near-optimal sparse-Blossom decoding, belief-matching, BP-OSD for LDPC/qLDPC codes, streaming/hybrid decoders, and high-throughput CPU/GPU batch decoding, with first-class Stim and Sinter integration. v0.6.6 is a critical-fix release that restores package importability after a total import failure affecting every published v0.6.5 wheel. It retains the full v0.6.4/v0.6.5 feature set. This whitepaper documents the full architecture, decoder catalogue, performance benchmarks against PyMatching (latency and logical error rate), throughput results, tail-latency analysis, and explicit accuracy/speed trade-offs across decoder families. All numbers are traceable to environment-stamped, content-hashed benchmark artifacts and are fully reproducible. The platform is designed for research, Monte Carlo validation, and Stim/Sinter-based workflows. It is not positioned as a universal replacement for PyMatching but offers measured advantages on repetition codes across tested distances and on surface codes at low-to-moderate distance.
No takes yet. Share an insight, caveat, or question.
Guillaume Besnard (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: