#rigetti

Rigetti Computing Signs $8.4 Million Quantum Partnership with C-DAC of India Rigetti Computing India Quantum Collaboration

Rigetti Computing India P L, a subsidiary of Nasdaq-listed Rigetti Computing, Inc., has placed a $8.4 million (£6.6 million) purchase order for a state-of-the-art quantum computer in India. The premier research and development organization under India's Ministry of Electronics and Information Technology, C-DAC, will receive the 108-qubit supercomputer.

The system is planned to be installed at C-DAC's Bengaluru center in the second half of 2026, marking a milestone in India's technology strategy. Technical Advance: 108-Qubit System The new system relies on Rigetti's chiplet-based design. This modular strategy may enable scaling quantum systems to extremely high qubit counts for fault-tolerant quantum computing and error correction. Rigetti uses superconducting qubits, unlike its competitors. This mode is preferred by the company due to its faster operations, obvious growth route, and maturity. The business believes Rigetti systems' gate speeds are 50 to 70 nanoseconds, 1,000 times faster than ion traps or neutral atoms. The Cepheus quantum computer being supplied to India has 108 qubits. These devices with 36 to 108 qubits use the company's unique control electronics and multi-chip processor technologies. Rigetti launched the Cepheus-1-36Q in 2025, the first multi-chip quantum processor, by tiling four 9-qubit chiplets. Strategic Supercomputing Infrastructure Integration in India The purchase is strategic integration into India's HPC architecture, not merely hardware. According to Rigetti CEO Dr. Subodh Kulkarni, the open and flexible architecture was designed to facilitate hybrid classical-quantum supercomputing. C-DAC's Hybrid HPC-Quantum Mission relies on this integration. By bringing the system on-site, C-DAC wants to provide its scientific and industry partners real R&D. Mission focus: Developing quantum accelerators. Establishing a national quantum computing reference center. Quantum communication and software stack and middleware development. Using hybrid HPC-quantum solutions in national scientific and industrial domains.

Explaining AI-QAN

Scientists have closed the gap between accurate prediction models and essential model interpretability, advancing astrophysical machine learning. This unique technology, the Explainable AI-enhanced Quantum Adversarial Network (XAI-QAN), combines XAI and classical deep learning with quantum-inspired neural networks.

The study, led by Sathwik Narkedimilli from Télécom Paris and Institut Polytechnique de Paris, aims to model galactic velocity dispersion (logσ e). Collaborators include Oracle Financial Services Software Ltd., the Indian Institute of Information Technology Dharwad, and others. Using complex MaNGA survey data.

The Astrophysics Interpretability Imperative

Velocity dispersion shows galaxies' history, mass distribution, and dynamics. Predicting this quantity is necessary to characterise the complex interplay between a galaxy's structure and dynamic activity. The 11 variables of the MaNGA dataset—morphological categorisation, effective radius (logR e), and gradients in star age and metallicity—provide a solid framework for this analysis.

However, Convolutional Neural Networks (CNNs) and Bayesian Neural Networks (BNNs) used in this sector are often 'black boxes'. The physical importance of these models' predictions is difficult to interpret, therefore there is a research gap despite their accuracy. XAI-QAN shrinks this gap using transparent and effective solutions. Cybersecurity, finance, and healthcare require precision and interpretability, thus this combination helps.

XAI-QAN increases quantum adversarial network An Evaluator Model and a Hybrid Quantum Neural Network (QNN) form the "Vanilla Model," the system's core.

Model 1: The Hybrid QNN uses quantum computing and deep learning to improve feature extraction. Initially, several completely connected layers preprocess the input data. This processed data is transmitted into a quantum layer using parameterised quantum circuits and R y and R x rotations on a 4-qubit system to determine the expected values of a Hamiltonian. Final prediction (y^) is obtained by running the quantum output via a sigmoid activation function. The model uses CUDA-Quantum on A800 NVIDIA GPUs and the parameter-shift algorithm for gradient propagation.

Adversarial Evaluator (Model 2) and LIME Integration: The Evaluator Model, an adversarial feedback feedforward neural network, complements the QNN. Importantly, the Evaluator's input vector (x ′) combines the original features (x), QNN's prediction (y^), and matching LIME explanations (E(x)). Local explanations from LIME explain how input properties affect prediction. In addition to calculating a feedback loss term, the Evaluator Model evaluates this combined output. Optimise it with a combined loss function (LQNN) that includes the QNN's direct prediction error (LM1).

MSE weighted by evaluator feedback loss (α⋅Levaluator). This ongoing adversarial loop ensures that the forecast and XAI-generated explanation match the target (y) for performance and interpretability consistency. The Adam optimiser and Mean Squared Error (MSE) loss are used to train the QNN and Evaluator across 10 epochs with a feedback weighting coefficient (α) of 0.5.

Experimental Results Support XAI-QAN Robustness

Empirical tests showed the Vanilla model was most accurate. It had a R 2 of 0.59, MAE of 0.21, MSE of 0.071, and RMSE of 0.27. This performance was comparable or better than the Quantum Self-Supervised, Q-GAN-1, and Q-GAN-2 models.

The training loss curves show that the adversarial guidance mechanism works: the QNN's MSE loss converged between 0.075 and 0.071, while the Evaluator Model's MSE loss steadily decreased from 0.28 to 0.20 over 10 epochs, demonstrating its improved ability to evaluate and direct the primary QNN model.

The Vanilla model had the lowest ECE of 0.015 and ACE of 0.012, indicating great calibration and dependability. Research showed that the model's prediction intervals were somewhat lower than necessary, resulting in a slight uncertainty estimate undercoverage, but this shows that the model's predictions are properly calibrated.

Interpretable Findings and Resource Efficiency

An advantage of XAI integration is the ability to appropriately comprehend feature contributions. According to the virial relation, feature perturbation analysis demonstrated that log M1/2 (enclosed mass) dominates all models' predictive capacity, accounting for 83–87% of predictions. Metallicity ([Z/H]) and half-light radius (log Re) were secondary factors at 5–7%, although star population gradients had little effect.

Total design efficiency is very low with 46,723 trainable parameters (0.18 MB). With the lowest inference latency (1.8 ms/sample) and energy consumption (0.027 J/sample), the Vanilla model showed its hybrid structure is efficient. Resource profiling showed that raising the qubit count from 1 to 4 improved performance across all models, confirming the benefit of deeper quantum feature spaces for expressive data representations.

Future Prospects and Limits

Quantum-inspired and deep learning methods can create competitive and interpretable predictive models, according to the study. It shows that evaluator feedback, classical layers, and the quantum layer are all necessary for top performance because removing one substantially reduced performance.

RIGetti Crosses Quantum Chasm: Third Quarter 2025 Highlights Key Contract Wins: Architectural and Software Milestones

This pioneer in hybrid quantum-classical computing started a discourse about Rigetti Computing, Inc.'s shift from research to scalable commercial deployments. It announced that it would have an investor conference call on November 11 at 8:30 a.m. Eastern Time (ET) and release its third-quarter 2025 financial results on November 10, 2025, after market close. The findings and corporate updates will be the main focus of this conference call by executives.

Investors and market observers have one of the greatest insights yet of whether Rigetti is moving beyond the “R&D phase” with the Third Quarter 2025 report and call.

Financial data is fictional.

Remember that the sources did not give Rigetti Computing's Q3 2025 financial statistics. So, the specific financial statistics below are fake and only for illustration. But the following technological and industry analysis are based on quantum computing trends and advancements.

Managing Execution and Budget

In the fictitious scenario, the quarterly release on October 29, 2025, portrayed aggressive investment in next-generation hardware design and software tools. This approach to quantum ecosystem accessibility and scalability is calculated.

The company's third-quarter sales of $4.7 million exceeded the analyst average of 98 million in cash and cash equivalents. Management believes this allows for expansion at technological milestones in 2027.

The report attempted to highlight that the company is carefully reducing its burn rate despite deep-tech industry losses during R&D. The management maintained that quantum utility, where quantum technologies outperform traditional systems, is the main element affecting long-term shareholder value.

Progress: Contracts and Partnership

In addition to the Third Quarter 2025 financials, Rigetti has made several public announcements in 2025, showcasing active cooperation and government and research contracts.

Achievements include:

The company received $5.7 million in purchase orders for two Novera quantum computing devices.

Rigetti got a $5.8 million Air Force Research Laboratory contract to enhance superconducting quantum networking with QphoX.

An MOU was signed with India's Centre for Development of Advanced Computing (C-DAC) to study hybrid quantum computing systems. Rigetti now supports NVIDIA NVQLink, an open quantum computing and AI supercomputing platform.

Despite focussing on R&D, the company is aggressively building strategic relationships.

Quantum Utility Race: Architectural Changes

In Q3 2025, the operational narrative focused on scaling systems beyond Noisy Intermediate-Scale Quantum (NISQ) technology. Technology focused on modular quantum processor design.

Such modular architectures may aid superconducting circuits, which struggle to connect many qubits with zero error. A reconfigurable router by Rigetti aims to enable all-to-all connectivity among smaller, easier-to-manage processor units. This strategy reduces faults from longer gate sequences and provides computational flexibility. Continuously improving control mechanisms is necessary to achieve low-error logical processes.

Reaching Out: Quantum Software and Accessibility

Rigetti focused on Quantum Software Engineering after realising hardware development is only half the battle. To go beyond the moment when significant physics knowledge is needed, quantum computing needs classical software developers to use it.

Software frameworks that are stable and hardware-independent are Rigetti's focus. Systems that translate classical issue requirements into quantum-executable code while maintaining methodological rigour are the goal. Rigetti hopes to automate problem classification, circuit construction, and hardware recommendation based on realism, cost, and runtime to substantially reduce the time needed to build industry-relevant applications.

quantum cloud security and reliability

As computing moves to the cloud, code and data security challenges arise. In his outlook, Rigetti explored QTEEs. Quantum circuits on remote, cloud-based computers need reliable hardware to hide and protect them. This requires new security against “side-channel” risks such qubit reset attacks that steal data between user programs.

Prospects and Investor Focus

Q3 results will be released after market hours on November 10, making the investor conference call on November 11 crucial for management discussion and recommendations.

Investors are still concerned about Third Quarter 2025 revenue trends, whether previously low operating losses and revenues will persist, cash reserves in relation to the burn rate, and technical milestones like lowering error rates and introducing new systems.

Investors and analysts will likely ask about scaling, cost control, timetables for larger-qubit systems, and Rigetti's plan for changing from grant/contract-driven revenue to recurring commercial models.

Conclusion,

Rigetti Computing Buys Two Novera Quantum Processors for $5.7 Million Today, Rigetti Computing, Inc., a leader in full-stack quantum-classical computing, announced purchase orders for two of its 9-qubit Novera quantum computing systems worth over $5.7 million. The company plans to deliver both systems in the first half of 2026. This revelation suggests an increasing need for on-premises quantum technology as organisations globally gain quantum experience.

The transaction shows how scientists and IT corporations use Rigetti hardware for quantum research. National labs and researchers throughout the world continue to pick and trust the Novera QPU to develop quantum computing technology R&D', said Rigetti CEO Dr. Subodh Kulkarni. I love seeing on-premises quantum computing technologies grow as the industry does.

Diverse Customers Drive Quantum Innovation

The two systems are for customers with different but equally ambitious quantum computing goals. One significant Asian technology manufacturer bought a Novera system. The system will be important to this company's quantum computing development. The company plans to benchmark and test its patented quantum computing technology using the Novera system to demonstrate its value in a competitive R&D setting. The second system will go to a California AI/physics startup. Using Novera technology, this company will do fundamental quantum error correction and quantum hardware research. Users can modify both systems to add qubits for more complex calculations and research. Inside Novera Quantum Processor Rigetti's Novera Quantum Processor is a complete on-premises system for practical, high-performance quantum computing research and development. The system has integrated, cutting-edge components. Its core is a Rigetti Ankaa-class 9-qubit quantum processor. The square lattice of qubits and tunable couplers in this processor enable high-fidelity two-qubit operations, which form complicated quantum algorithms. Subspace temperatures are needed to operate these sensitive quantum devices. The Novera system's appropriate dilution refrigerator keeps qubits cool and vacuum-sealed, shielding them from outside noise. The package includes Rigetti's unique control systems for qubit state control, calibration, and readout.

The Novera system was designed to help quantum computing research in key areas like:

Rigetti Computing Launches 84-Qubit Ankaa 3 System with 99.5% Median Two-Qubit Gate Fidelity. Rigetti tops the market with its revolutionary hardware revamp that boosts performance, scalability, and authenticity.

Features

Better Qubit Integrity

The median two-qubit gate integrity of 99.5% in Ankaa 3 cuts error rates in half compared to previous systems. Its median fidelity for fSim gates is 99.5% and for iSWAP gates is 99.0%, which is crucial for complex quantum algorithms like random circuit sampling.

Innovational Cryogenic Architecture

A novel cryogenic hardware design reduces metal content at the refrigerator's coldest point. This innovation improves heat management, magnetic/environmental shielding, qubit cost, and system performance.

Improved Qubit Chip

Rigetti redesigned its qubit for coherence. In collaboration with Fermilab-led SQMS, a new metal deposition approach for qubit circuitry resulted in increased T₁ baseline and longer qubit lifetimes. The circuit configuration was designed to reduce qubit losses.

Josephson Junctions ABAA Method

Rigetti's Alternating-Bias Assisted Annealing (ABAA) technique creates Josephson junctions in the Ankaa 3 chip. This method increases two-qubit gate performance and yield by accurately targeting qubit frequency, increasing fidelity.

Scalable architecture

Ankaa-3 retains Rigetti's scalable chip design with 3D signal transmission. This design supports the integration of many qubit devices, making larger quantum systems easier to create. Rigetti plans to release a system with more than 100 qubits by the end of 2025 and a 36-qubit system constructed on four 9-qubit chips tiled together by the middle of 2025 to reduce mistake rates by twofold.

Meaning of Ankaa 3

It Errs Less:

Ankaa 3 is more accurate than previous systems.

Quantum computations with fewer errors are essential for addressing difficult problems.

It Helps Solve Bigger Issues:

Its power (84 qubits) lets it do more demanding tasks.

This helps Artificial intelligence, finance, climate research, and medicine.

A Big Step Forward:

Researchers aim to build computers that outperform supercomputers.

With Ankaa-3, we're approaching quantum advantage.

Smarter Construction:

Future expansion is easier and more efficient with the design. Like Lego parts that can be connected to make a bigger object.

Keeps Cool and Saves Energy:

Superior cooling systems make Ankaa 3 cheaper to operate.

This is important because quantum computers operate at very low temperatures.

See also Norma, Rigetti 84-Qubit Quantum Cloud Service to South Korea

More People Can Use It:

Ankaa-3 works without a quantum computer.

It is available online to researchers and enterprises globally via cloud platforms like Microsoft Azure and Amazon Braket.

It uses better technology:

Rigetti built the system's primary parts in an innovative way.

The machine is more reliable and easier to build.

Component of a Bigger Scheme:

Rigetti wants more advanced and expansive systems.

From Ankaa 3, they plan to build systems with over 100 qubits.

Briefly:

Ankaa-3 is faster, smarter, and precise.

Researchers and developers can use quantum computers better.

It's a big step towards powerful computers that can tackle real-world problems.

Prospects and availability

Partner access to Ankaa-3 is via Rigetti Quantum Cloud Services (QCS). It is scheduled to be available on Microsoft Azure and Amazon Braket in the first quarter of 2025, expanded to researchers and developers worldwide.

With no debt and $225 million in cash, cash equivalents, and available-for-sale investments, Rigetti is well-positioned to advance quantum computing. To achieve fault-tolerant quantum computing, the company is enhancing system performance and scalability.

In conclusion

Mark Zuckerberg ve Nvidia CEO’su Jensen Huang, kuantum teknolojisinin kullanım için henüz hazır olmadığını belirtti. Bu açıklamalar, sektördeki büyük hisselerin değer kaybetmesine yol açtı. Zuckerberg ve Huang’ın Açıklamaları Piyasayı Sarstı Meta CEO’su Mark Zuckerberg, kuantum bilgisayarların kullanıma hazır hale gelmesinin yıllar alacağını söyledi. Nvidia CEO’su Jensen Huang da benzer bir yorum…

Quantum Computing, D-Wave Quantum, and Rigetti Stocks Plunge. Blame Nvidia’s CEO.

SEE MORE DELATIES CLICK ON LINK