A downloadable tool

Download NowName your own price

QECTOR DECODER WORKBENCH (Linux Edition)

Version 3.5.1 · Backend: qector-decoder-v3==0.6.9

Professional desktop application for quantum error correction research, evaluation, and documentation on Linux.

  • 13 decoder algorithms
  • 9 code families (surface, toric, heavy-hex, bicycle qLDPC, …)
  • 47-tool Model Context Protocol (MCP) server for AI agents
  • Native Debian package (.deb) for Linux x86_64
  • Full GUI + command-line + headless MCP mode

Website:

  • https://www.qector.store

    Author: Guillaume Lessard / iD01t Productions · ORCID 0009-0000-3465-3753

    Package & Installation

    Package File: qector-workbench_3.5.1_amd64.deb

    Quick Start:

    1. Install the Debian package:

      sudo dpkg -i qector-workbench_3.5.1_amd64.deb

      (Alternatively: sudo apt install ./qector-workbench_3.5.1_amd64.deb)

      Launch the application:

    2. qector-workbench

      Launch the MCP Server (AI agents):

    3. qector-workbench --mcp

      System Requirements

    • OS: 64-bit Linux x86_64 (glibc >= 2.30; Ubuntu 20.04+, Debian 11+, Linux Mint 20+, Fedora 32+, openSUSE Leap 15.3+, Arch, etc.)
    • Runtime: Python 3.10+ and python3-pip (required for backend auto-provisioning)
    • Memory: 4 GB minimum (8 GB recommended)
    • Disk Space: ~150 MB
    • Network: Internet required only on first launch to download the backend from PyPI (subsequent runs operate 100% offline)

    Key Features

    • Interactive GUI: 7 feature tabs (Code Explorer, Decoder Lab, Benchmark Suite, Batch & Streaming, Hardware & System, Diagnostics, Documentation Studio) plus a live colored log console.
    • 13 Decoders: Includes Union-Find, Fast Union-Find, Blossom (exact MWPM), Sparse Blossom, BP-OSD, Belief Matching, GNN Belief Matching, Hybrid Cascade, Auto, Predecoded, Lookup Table, and Auto-Router.
    • 9 Code Families: Covers graph-like codes (repetition, ring, rotated surface, unrotated surface, toric, heavy-hex, hypergraph product) and qLDPC codes (bicycle, bivariate bicycle).
    • Publication-Grade Documentation: Multi-format exports (MD, JSON, HTML, LaTeX, PDF, SVG) with embedded provenance metadata.
    • 47-Tool MCP Server: Native stdio JSON-RPC 2.0 integration for Claude, GPT, Copilot, or custom AI agents.
    • Automatic Provisioning & Self-Healing: Automatically fetches qector-decoder-v3 from PyPI into an isolated per-user site directory on first run and auto-recovers if corrupted.

    Licensing

    • Workbench (Application): Covered under EULA.txt. Royalty-free worldwide license for any purpose (including commercial) as long as embedded QECTOR notices are retained.
    • Backend (qector-decoder-v3): Separately licensed Rust/Python platform. Free for personal, academic, educational, and non-commercial research. Commercial use requires a paid license (a 60-day commercial evaluation is available at qector.store/pricing).

    Included Documentation

    Located in the installed package and docs/ folder:

    • README.txt: Full package overview
    • EULA.txt: End User License Agreement
    • QECTOR_User_Manual_Linux.pdf: Linux user manual
    • QECTOR_API_Reference.md / .pdf: Complete API reference specification
    • QECTOR_MCP_Integration_Guide.pdf: MCP server integration guide
    • QECTOR_Quick_Start_Guide.pdf: Quick start guide
    • QECTOR_LLM_Manual.json: Machine-readable operational manual for LLMs/agents

    Citation

    Lessard, Guillaume. QECTOR Decoder v3: Rust/Python Quantum Error Correction Decoding Platform. Version 0.6.9. 2026.

  • https://www.qector.store 
  • ORCID: 0009-0000-3465-3753


  • Related Links

    • More information

    Download

    Download NowName your own price

    Click download now to get access to the following files:

    qector-workbench_3.5.1_amd64.zip 677 kB

    Development log

    Leave a comment

    Log in with itch.io to leave a comment.