Qilimanjaro Releases QiliSDK 0.2.0 and Integrates NVIDIA CUDA-Q for Unified Hybrid Hardware Emulation

Barcelona-based quantum developer Qilimanjaro Quantum Tech has announced the release of QiliSDK 0.2.0, an open-source Python framework designed to unify the programming landscape for analog and digital quantum systems. The new version introduces a backend-agnostic API, allowing researchers to develop a single high-level codebase and easily switch execution between CPUs, GPUs, and physical Quantum Processing Units (QPUs). This abstraction layer aims to resolve the longstanding challenge of fragmented toolkits that typically restrict users to specific physical paradigms.

A major highlight of this release is the direct integration of NVIDIA CUDA-Q. By utilizing the CudaBackend class, users can leverage massive GPU parallelization to handle state-vector simulations beyond the 25-qubit limit, reaching 30 qubits and more on single-node classical supercomputers. This integration also optimizes multi-node tensor-network contractions, providing a significant boost in performance for hybrid quantum-classical workflows.

Additionally, version 0.2.0 introduces a comprehensive noise modeling engine supporting both Kraus and Lindblad representations, and adds native support for OpenQASM 3 and Microsoft's Quantum Intermediate Representation (QIR). The framework also includes specialized primitives for Quantum Reservoir Computing, facilitating more efficient analog hardware research. This update is part of a broader European effort to link NVIDIA-accelerated supercomputing centers, such as the Barcelona Supercomputing Center (BSC), with sovereign physical quantum infrastructure.

Source: quantumcomputingreport.com
Publication date: 23.06.2026
Author: Mohamed Abdel-Kareem