Postquantum Labs has launched QVM, a virtual machine designed to standardize quantum computing languages and pricing, currently live on the Aglais testnet with initial support for both D-Wave and IonQ systems. The new platform aims to address a critical portability problem in the field, where programs built for one system often cannot run on another. Microsoft’s Azure Quantum, for example, typically requires a $1,000 spend commitment before a user can run their first circuit.
“Oftentimes, when we put a student on a project, it takes them the whole semester just to learn how to program using the latest quantum paradigms,” says Dr. Terrence Mak of Monash University, believing QVM can significantly reduce that learning curve.
QVM Unifies Quantum and Classical Instruction Sets
Quip Network has launched the Quip Virtual Machine (QVM) with the aim of standardizing quantum operation costs across different hardware platforms. The system uses a “gas-based” pricing model to predict computation expenses before execution. This approach, according to the company, transforms quantum tests from research experiments into practical engineering tools by allowing for quantifiable value assessment.
The architecture of QVM addresses a critical issue in quantum computing: the lack of portability between systems. Quip Network’s approach allows developers to write programs in a single constraint-based language applicable to annealers, gate-based quantum systems and classical hardware, eliminating the need for machine-specific code.
Programs compiled through QVM deploy to the Quip Network, executing on connected quantum computers while classical machines verify the results, and developers can earn revenue each time their program runs, the company says. “We created the QVM because quantum computing is still fractured between competing instruction sets and assemblers, providing few easy onramps to outsiders,” explained a company spokesperson.
“Our virtual machine has three major goals: to unify and clarify pricing of quantum operations and memory across architectures, to distribute hybrid computation across available machines on our network, and to provide an easier abstraction for new quantum programmers that doesn’t require them to think directly in quantum gates or qubit layouts.” SiC Systems, a company specializing in chemical and biological manufacturing, is already exploring the potential of QVM to optimize complex processes.
The company’s focus on combinatorial optimization problems requires rapid solution exploration, a need QVM is designed to address, according to Quip Network. “At SiC Systems, we face a variety of combinatorial optimization problems, and time is of the essence,” a representative stated. “We want to explore as many solutions as possible in as short a time as possible.
This is particularly important in our physics-informed agent architecture to design, control and operate complex chemical, and biological manufacturing processes. Quip Network’s QVM can help us meet that requirement and better serve our customers.” Researchers at other institutions anticipate QVM will streamline the application of quantum methods to practical research, including work on the Lagrangian operator, by reducing the time spent learning specific quantum programming tools.
Using a pay-as-you-go ‘gas-based’ cost model is the missing piece that demonstrates quantum computing is commercially viable and practical. When you can predict exactly what a computation will cost before you run it, quantum tests that began as a research experiment become a real tool for engineering because you can quantify exactly how much value it can bring to your products.
Constraint-Based Language Enables Hybrid Computation
The new Quip Virtual Machine (QVM) addresses a significant economic hurdle in quantum computing by offering predictable costs before execution, a feature previously unavailable across different hardware platforms. Unlike pricing structures, which typically require roughly $1,000 in spend before a first circuit can run, QVM uses a “gas-based” model to quantify computational expense. This system calculates cost based on qubit space, execution time and entanglement usage, allowing developers to budget accurately for quantum workloads.
Developers write programs once, and the QVM compiles them into instructions suitable for the available hardware, eliminating the need for separate codebases for each platform. This distributed approach allows for hybrid computation by using the strengths of both quantum and classical processing. Developers can publicly publish programs and earn revenue each time they run, creating an incentive for code sharing and collaboration. This initial release supports both D-Wave and IonQ quantum systems alongside classical computers, representing a first step toward a unified representation for all of computing.
We created the QVM because quantum computing is still fractured between competing instruction sets and assemblers, providing few easy onramps to outsiders. Our virtual machine has three major goals: to unify and clarify pricing of quantum operations and memory across architectures, to distribute hybrid computation across available machines on our network, and to provide an easier abstraction for new quantum programmers that doesn’t require them to think directly in quantum gates or qubit layouts.
QVM Gas Pricing Predicts Quantum Operation Costs
The barrier to entry for quantum computing research extends beyond access to hardware. A fragmented pricing structure adds significant overhead. QVM’s architecture aims to reduce the time spent on tooling, allowing researchers to focus on their core questions. Dr. Terrence Mak of Monash University notes that students often dedicate an entire semester to learning quantum programming paradigms before beginning their research, a timeframe Quip hopes to shorten.
Programs written in this language compile into instructions suitable for available hardware and deploy to the Quip Network, a distributed system using both quantum and classical computers. The Quip Network’s operation is facilitated by QuipSwap, a settlement layer live since September that handles payments with quantum-resistant signatures. This approach contrasts with vendor-specific pricing like that of IBM Quantum, which applies only to its own hardware.
Working with Quip’s QVM can help researchers and students spend less time learning specific quantum programming tools and more time applying quantum methods to practical research problems, including our work on the Lagrangian operator.
Verification Tiers Ensure Result Trustworthiness
Every computation on the Quip Network undergoes a rigorous three-tiered verification process, designed to establish trust in results regardless of the underlying hardware. This system begins with classical simulation, where quantum outputs are cross-checked against predictions from conventional computers, providing an initial layer of validation. Independent benchmarking then subjects those outputs to further scrutiny. Classical providers run the same computations on their own hardware to confirm the findings.
Finally, cryptographic attestation binds each result to the specific machine that generated it, creating an immutable record of provenance. This multi-layered approach addresses a critical challenge in early quantum computing: ensuring the reliability of outputs when direct verification is difficult. Should a result be disputed, the network doesn’t rely on the operator’s claims, but instead settles the disagreement against the pre-recorded benchmarks. This system is not merely theoretical; during a live demonstration at Quantum Tech World, the QVM was used to solve 1,930 portfolio optimization problems, pitting quantum and classical solvers against each other.
The quantum solver either matched or outperformed the classical solver in every race, with a median improvement of 2.56% in cases where differences exceeded 0.5%. Soheil Mansouri, Co-Founder and Chief Science Officer, explained that this is facilitated by the QVM, which is the programming layer, allowing developers to access annealers, gate-based systems and classical computers through a single, unified language. The network’s settlement layer, QuipSwap, launched in September, handles payments and ensures secure transactions with quantum-resistant signatures, completing the ecosystem.
Oftentimes, when we put a student on a project, it takes them the whole semester just to learn how to program using the latest quantum paradigms. As we evaluate Quip’s virtual machine, we believe it can speed up student training and give them more time to do real research instead of spending cycles learning a specific programming language or how to program quantum circuits.
See today’s quantum computing news on Quantum Zeitgeist for the latest breakthroughs in qubits, hardware, algorithms, and industry deals.
