xi-zhao/openquantum
Developer toolsThis page was translated automatically. Technical details may differ from the original. View the original
Open-source quantum Agent workspace with a desktop client, Web UI, messaging, Qiskit/MCP tools, and scientific validation
Overview
Chinese Simplified · English · Japanese · 한국어 · Español · Français · Deutsch · Português · Русский · العربية OpenQuantum connects professional quantum software, research methods, and complete applications to your problems. You can initiate calculations using natural language, check the results returned by tools, prepare courses on Quantum Learning Pass, or integrate your own algorithms and applications. From a learner's first experiment, to comparing researchers' methods, and then to developer application integration, all can start from supported tasks. Why Choose · Browsing Capabilities · Getting Started · Usage and Results · Development and Scaling · Routes and RSI · Participation in Construction · Open Source Ecosystem OpenQuantum's work is to bring domain methods, agent execution, and open tools to specific tasks; Its value doesn't have to wait until real QPUs gain an advantage; local computing, teaching experiments, and method validation are also practical uses. You don't have to learn every SDK before starting a supported task: Skill provides working methods, Agents call professional tools, and the workbench retains input and results. We hope to reduce repeated configuration, interface learning, and process building, allowing learners to get hands-on and researchers to focus more on hypothesis, designing comparisons, and judging results, rather than replacing those judgments. PyZX circuit optimization, QCEC equivalence check, and noise simulation each answer different questions. Using them for the same research objective allows us to further ask whether "the ideal circuit is simpler and has better noise performance." We continue to build this scientifically meaningful combination: aligning models, units, positions, and budgets to understand when methods are applicable and when they fail; Interface connection itself does not mean the conclusion is valid. The current scope of each tool is shown in computation and resource allocation. Common steps, calculation results, and applicable independent checks provide a foundation for review and subsequent use. We hope to further accumulate conditional method comparisons and evidence of failure: not only knowing a successful calculation once but also understanding why it applies and when to switch routes. Well-founded negative results are also valuable, but procedural errors do not equal scientific refutation. Complete scientific acceptance is currently provided within the scope of corresponding capabilities, and experience compilation does not mean the system has learned automatically.
Overview (machine translated)
Chinese Simplified · English · Japanese · 한국어 · Español · Français · Deutsch · Português · Русский · العربية
OpenQuantum co-builds trusted quantum research intelligence.
Pooling the professional accumulation of quantum research into one OpenQuantum platform
OpenQuantum is an open-source quantum research AI platform that connects outstanding software, algorithms, and publicly available research results from the open-source community. It allows AI to analyze, select methods, perform calculations, and verify results around your research questions, transforming professional expertise into practical research capabilities.
Start from a research question, a paper, or an existing program to let AI participate in scientific exploration. Retain procedures, parameters, and verification records, turning one research into the starting point for the next.
Choose the installation entry that suits you best, and both will reuse the same domain capabilities: Codex plugin connects Quantum Skill and Science Tool into Codex’s native dialogue, with side entry for managing capability switches, MCP connections, and tokens; Desktop includes a research workbench and a standalone Quantum Learning Pass application, with model services configurable by default.
Install Codex Plugin · Install Desktop · First, look at the calculation and results · Visit the official website
Let’s first look at the test case: PennyLane Pauli speeds up full call calls to computation by 5.09 times; QuTiP Wigner function speeds up full call computation by 5.69 times, all with performance comparison, result verification, and rerun records.
Current version 0.8.4: Desktop download and release instructions · Codex plugin download and release instructions. Both use the same product version number.
- Easier to Find Skills: Tools, MCP, and Skills are grouped by 8 types of uses, supporting search and status filtering.
- Unified brand entry: Mac
OpenQuantum.appand WindowsOpenQuantum.lnkuse the OpenQuantum icon; Retain the original DSH Desktop interface, Quantum Learning Pass, and update portal.
Why Choose · Optimized Tests · Browsing Capabilities · Codex plugin · Get started · Usage and Results · Development and Expansion · Route and RSI · Participation in Construction · Open source ecosystem
Keep the familiar original Desktop interface; Add your own model service in the settings to start a new session.
View the classification settings for Tools, MCP, and Skills
Expand the required grouping by purpose, then narrow it down using search or enable status. Tools are used to view callable actions, MCP is used to manage service connections, and Skills are used to select task methods and guides.
The image shows the actual interface of the classification function verified in the original client on 2026-10-13. Screenshot source is shown in the image description. Enabling configuration does not mean the service is online.
Why Choose OpenQuantum
This page was translated automatically. Technical details may differ from the original.
Recommended Tools
Try a different keyword or remove a filter.
Install
npx skillfish add xi-zhao/openquantum