CUDA-Q

<!-- Release notes generated using configuration in .github/release.yml at releases/v0.16.0 -->

What's Changed

Features and Enhancements 🎉

• Support disconnected device topologies in the qubit mapper by @khalatepradnya in https://github.com/NVIDIA/cuda-quantum/pull/4824
• Add Python 3.14 support by @mitchdz in https://github.com/NVIDIA/cuda-quantum/pull/4670
• Expanded mapping support for control flow by @atgeller in https://github.com/NVIDIA/cuda-quantum/pull/4838
• Support for static Rotation Gate Synthesis (Clifford + T) by @sacpis in https://github.com/NVIDIA/cuda-quantum/pull/4508
• Add the ability to disable all quantum optimizations by @schweitzpgi in https://github.com/NVIDIA/cuda-quantum/pull/5057
• Introduce atomic quantum regions by @khalatepradnya in https://github.com/NVIDIA/cuda-quantum/pull/5275
• Adding CUDA-Q Logical by @bettinaheim in https://github.com/NVIDIA/cuda-quantum/pull/5422

Bug Fixes 🐛

• Fix multi-qubit noise broadcast in Stim MSM mode by @bmhowe23 in https://github.com/NVIDIA/cuda-quantum/pull/4731
• [python] Serialize MLIR pass manager runs for async compilation by @khalatepradnya in https://github.com/NVIDIA/cuda-quantum/pull/4771
• [braket] Fix the basis gate set for decomposition so that unsupported gates such as sdg and tdg are not emitted by @khalatepradnya in https://github.com/NVIDIA/cuda-quantum/pull/4787
• Fix cudaq.evolve with shared Hamiltonian and batched initial states by @huaweil-nv in https://github.com/NVIDIA/cuda-quantum/pull/4835
• Support individual-qubit operand in exp_pauli by @1tnguyen in https://github.com/NVIDIA/cuda-quantum/pull/4852
• Fix mixed int/float arithmetic promotion by @huaweil-nv in https://github.com/NVIDIA/cuda-quantum/pull/4854
• Fix dataclass kernel argument containing list raises std::bad_cast by @atgeller in https://github.com/NVIDIA/cuda-quantum/pull/4860
• Correctly handle early return in Python by @1tnguyen in https://github.com/NVIDIA/cuda-quantum/pull/4864
• [python] Fix translation of return-typed kernels by @khalatepradnya in https://github.com/NVIDIA/cuda-quantum/pull/4858
• Qubit ordering problem on IQM backends #4621 by @iqm-bhoffmann in https://github.com/NVIDIA/cuda-quantum/pull/4818
• [runtime] Preserve measurement handles across nested JIT calls by @khalatepradnya in https://github.com/NVIDIA/cuda-quantum/pull/4861
• Fix type inference of captured arguments by @sacpis in https://github.com/NVIDIA/cuda-quantum/pull/4871
• Fix cudaq.run crash when remote server returns fewer shots than requested by @adityabagchi24 in https://github.com/NVIDIA/cuda-quantum/pull/4883
• Fix odd-electron UCCSD virtual orbital indices by @khalatepradnya in https://github.com/NVIDIA/cuda-quantum/pull/4893
• Implementing -MD, -MMD, -MT and -MF in nvqpp by @Renaud-K in https://github.com/NVIDIA/cuda-quantum/pull/4929
• Add support for empty state vectors in statevector simulator by @mitchdz in https://github.com/NVIDIA/cuda-quantum/pull/4947
• Enable qubit reuse in cudaq::run by @anpaz in https://github.com/NVIDIA/cuda-quantum/pull/4945
• Add an owned single-flight compiled-module cache by @khalatepradnya in https://github.com/NVIDIA/cuda-quantum/pull/4998
• Fix distributed explicit-state initialization in dynamic backend by @huaweil-nv in https://github.com/NVIDIA/cuda-quantum/pull/4988
• Fixing cudaq-translate crash on custom operations with -fno-array-conversion by @sacpis in https://github.com/NVIDIA/cuda-quantum/pull/5010
• Fix segfault crash when there are pending tasks in the QPU execution queue by @1tnguyen in https://github.com/NVIDIA/cuda-quantum/pull/5030
• Fix density matrix layout across QPP and dynamics when states are created from complex_matrix by @huaweil-nv in https://github.com/NVIDIA/cuda-quantum/pull/5053
• Fixing transposed closed-system propagator in cudaq.contrib by @sacpis in https://github.com/NVIDIA/cuda-quantum/pull/5093
• Preserving state layout metadata when splitting a batched state by @sacpis in https://github.com/NVIDIA/cuda-quantum/pull/5115
• Fix aliasing issues for quantum vectors by @schweitzpgi in https://github.com/NVIDIA/cuda-quantum/pull/5126
• Minor CMake fixes by @lmondada in https://github.com/NVIDIA/cuda-quantum/pull/5169
• Fail decomposition when no basis is supplied by @EvanGruhlkey in https://github.com/NVIDIA/cuda-quantum/pull/5158
• Selectively unroll loops with dynamic pauli-word indexing by @1tnguyen in https://github.com/NVIDIA/cuda-quantum/pull/5191
• Fixup qubit ordering problem on IQM backends by @iqm-bhoffmann in https://github.com/NVIDIA/cuda-quantum/pull/5074
• Fixing nested-loop adjoint by @sacpis in https://github.com/NVIDIA/cuda-quantum/pull/5233
• Fix MemToReg live-ins for cc.scope by @khalatepradnya in https://github.com/NVIDIA/cuda-quantum/pull/5238
• Fix Resource Estimation on value semantics by @lmondada in https://github.com/NVIDIA/cuda-quantum/pull/5148
• Fix unwind threading bug by @schweitzpgi in https://github.com/NVIDIA/cuda-quantum/pull/5246
• Fix exit block collection bug in memtoreg by @schweitzpgi in https://github.com/NVIDIA/cuda-quantum/pull/5256
• Fix a bug when initializer expression in missing in C++ for-loops by @schweitzpgi in https://github.com/NVIDIA/cuda-quantum/pull/5265
• Fix some bugs in handling of the Python for-else construct by @schweitzpgi in https://github.com/NVIDIA/cuda-quantum/pull/5282
• Ensure qubits in Python loops and ifs get freed by @sacpis in https://github.com/NVIDIA/cuda-quantum/pull/5292
• Fix some memtoreg issues by @schweitzpgi in https://github.com/NVIDIA/cuda-quantum/pull/5293

Breaking Changes 🛠

• realtime: self-relaunching device-graph scheduler for unbounded device-side dispatch by @cketcham2333 in https://github.com/NVIDIA/cuda-quantum/pull/4783
• Updating cuquantum to 26.06.0 by @sacpis in https://github.com/NVIDIA/cuda-quantum/pull/4815
• [realtime] Split graph launch dispatch from the host ring loop by @boschmitt in https://github.com/NVIDIA/cuda-quantum/pull/4770
• Removing deprecated SpinOp operations by @anpaz in https://github.com/NVIDIA/cuda-quantum/pull/4833
• Pluggable transport providers for cudaq-realtime by @bmhowe23 in https://github.com/NVIDIA/cuda-quantum/pull/4915
• Rename of the IR type stdvec by @schweitzpgi in https://github.com/NVIDIA/cuda-quantum/pull/5089

Documentation Updates ✏️

• [docs] Updated docs for stim::ErrorAnalyzer::circuit_to_detector_error_model flags in dem_from_kernel by @eliotheinrich in https://github.com/NVIDIA/cuda-quantum/pull/4795
• [docs] Document m2d/m2o measurement matrix output for dem_from_kernel by @bmhowe23 in https://github.com/NVIDIA/cuda-quantum/pull/4839
• Fix operator examples in docs by @mitchdz in https://github.com/NVIDIA/cuda-quantum/pull/4376
• Add CUDA-Q compiler develop documentation section with dialect references by @taalexander in https://github.com/NVIDIA/cuda-quantum/pull/4930
• Rename hololink GPU-side transceiver software by @cketcham2333 in https://github.com/NVIDIA/cuda-quantum/pull/5006
• Update the statevector simulators to leverage the latest improvements in cuquantum by @1tnguyen in https://github.com/NVIDIA/cuda-quantum/pull/5025
• Add documentation for developing the Cuda-Q MLIR compiler by @taalexander in https://github.com/NVIDIA/cuda-quantum/pull/4938
• Adding qbraid examples to the doc by @sacpis in https://github.com/NVIDIA/cuda-quantum/pull/5078
• Add cudaq-devel wheel by @lmondada in https://github.com/NVIDIA/cuda-quantum/pull/5087
• Adding qbraid to hw providers by @sacpis in https://github.com/NVIDIA/cuda-quantum/pull/5120
• Adding --installpath option to the realtime installer by @Renaud-K in https://github.com/NVIDIA/cuda-quantum/pull/5134
• Support scopes in qubit mapping by @taalexander in https://github.com/NVIDIA/cuda-quantum/pull/5124
• Add link to pulse research preview by @lmondada in https://github.com/NVIDIA/cuda-quantum/pull/5255
• Update documentation for atomic quantum regions by @khalatepradnya in https://github.com/NVIDIA/cuda-quantum/pull/5321

Other Changes

• Enable HOST_CALL entries in device_call host_dispatch channel by @vedika-saravanan in https://github.com/NVIDIA/cuda-quantum/pull/4729
• [unitaryhack] Add closed- and open-system dynamics propagators by @friedsam in https://github.com/NVIDIA/cuda-quantum/pull/4672
• Add shared-ring realtime dispatch routing support by @cketcham2333 in https://github.com/NVIDIA/cuda-quantum/pull/4712
• [unitaryhack] Support unitary synthesis for 3+ qubit custom operations (QSD) by @thedaemon-wizard in https://github.com/NVIDIA/cuda-quantum/pull/4693
• Add two quantum embeddings by @simsaidan in https://github.com/NVIDIA/cuda-quantum/pull/4667
• [unitaryhack] Topology-aware initial placement for the qubit-mapping pass by @border-b in https://github.com/NVIDIA/cuda-quantum/pull/4678
• Add kernel math function support by @LubuSeb in https://github.com/NVIDIA/cuda-quantum/pull/4680
• [macOS] Surface simulator kernel exceptions across the macOS JIT boundary by @mitchdz in https://github.com/NVIDIA/cuda-quantum/pull/4777
• Implement conditional loop unrolling for value-semantics pipelines by @1tnguyen in https://github.com/NVIDIA/cuda-quantum/pull/4784
• Accept np.floor in Python kernels by @aryanputta in https://github.com/NVIDIA/cuda-quantum/pull/4785
• Support trivial control flow in Mapping by @atgeller in https://github.com/NVIDIA/cuda-quantum/pull/4733
• Add support for UDP transport in cudaq-realtime by @bmhowe23 in https://github.com/NVIDIA/cuda-quantum/pull/4877
• Add unified CPU dispatch loop to cudaq-realtime by @boschmitt in https://github.com/NVIDIA/cuda-quantum/pull/4869
• Deal with disconnected-island packing in qubit mapping by @Adithyaphani in https://github.com/NVIDIA/cuda-quantum/pull/4885
• Avoid repeated scans for unbound measurement handles by @khalatepradnya in https://github.com/NVIDIA/cuda-quantum/pull/4937
• Make Python AST bridge terminator queries constant time by @khalatepradnya in https://github.com/NVIDIA/cuda-quantum/pull/4936
• Reuse dominance-safe fixed qvector references by @khalatepradnya in https://github.com/NVIDIA/cuda-quantum/pull/4943
• External backends Plugin Framework by @anpaz in https://github.com/NVIDIA/cuda-quantum/pull/4754
• Fix and issue with compiled-module cache reuse by @khalatepradnya in https://github.com/NVIDIA/cuda-quantum/pull/4978
• Use shared libcudaqMLIR dependency everywhere by @lmondada in https://github.com/NVIDIA/cuda-quantum/pull/4928
• Static-link CUDA runtime in cudaq-realtime UDP bridge provider by @Renaud-K in https://github.com/NVIDIA/cuda-quantum/pull/4999
• [python] Reuse compiled modules across asynchronous launches by @khalatepradnya in https://github.com/NVIDIA/cuda-quantum/pull/5008
• [core] Update wire and cable to be SSI types. by @schweitzpgi in https://github.com/NVIDIA/cuda-quantum/pull/5033
• Adding a negligible rotation pruning pass by @sacpis in https://github.com/NVIDIA/cuda-quantum/pull/4992
• Enabling quantum optimizations on value semantics by default by @schweitzpgi in https://github.com/NVIDIA/cuda-quantum/pull/4993
• [Python] Materialize broadcast arguments once by @khalatepradnya in https://github.com/NVIDIA/cuda-quantum/pull/5069
• Minor CMake hardening, fixes & tweaks by @lmondada in https://github.com/NVIDIA/cuda-quantum/pull/5054
• Representation and optimization of global phases during compilation by @cabreraam in https://github.com/NVIDIA/cuda-quantum/pull/4976
• Accept global-register expectations in async Future by @tomers-qedma in https://github.com/NVIDIA/cuda-quantum/pull/5101
• Support Python math functions in addition to numpy functions by @1tnguyen in https://github.com/NVIDIA/cuda-quantum/pull/5130
• Expose dependency-aware T-depth resource metric by @taalexander in https://github.com/NVIDIA/cuda-quantum/pull/5145
• Migrate phase folding for value semantics by @atgeller in https://github.com/NVIDIA/cuda-quantum/pull/5021
• Computing CuDensityMatState overlap on the GPU by @sacpis in https://github.com/NVIDIA/cuda-quantum/pull/5184
• Overhaul of apply op specialization by @schweitzpgi in https://github.com/NVIDIA/cuda-quantum/pull/4940
• Adding set_function_entry to brige API by @Renaud-K in https://github.com/NVIDIA/cuda-quantum/pull/5155
• Support std::fmod in C++ QPU kernels by @Bryanzzy1 in https://github.com/NVIDIA/cuda-quantum/pull/5217
• Tune gridsynth and expose controls by @sacpis in https://github.com/NVIDIA/cuda-quantum/pull/5143
• [core] Add a new pass to convert control flow to dataflow. by @schweitzpgi in https://github.com/NVIDIA/cuda-quantum/pull/5239
• Support basic measurement feedback in qubit mapping by @1tnguyen in https://github.com/NVIDIA/cuda-quantum/pull/5235
• Ergonomic improvements for downstream packages by @lmondada in https://github.com/NVIDIA/cuda-quantum/pull/5253

New Contributors

• @friedsam made their first contribution in https://github.com/NVIDIA/cuda-quantum/pull/4672
• @thedaemon-wizard made their first contribution in https://github.com/NVIDIA/cuda-quantum/pull/4693
• @simsaidan made their first contribution in https://github.com/NVIDIA/cuda-quantum/pull/4667
• @border-b made their first contribution in https://github.com/NVIDIA/cuda-quantum/pull/4678
• @LubuSeb made their first contribution in https://github.com/NVIDIA/cuda-quantum/pull/4680
• @Adithyaphani made their first contribution in https://github.com/NVIDIA/cuda-quantum/pull/4790
• @adityabagchi24 made their first contribution in https://github.com/NVIDIA/cuda-quantum/pull/4883
• @tomers-qedma made their first contribution in https://github.com/NVIDIA/cuda-quantum/pull/5101
• @EvanGruhlkey made their first contribution in https://github.com/NVIDIA/cuda-quantum/pull/5158
• @Bryanzzy1 made their first contribution in https://github.com/NVIDIA/cuda-quantum/pull/5217

Full Changelog: https://github.com/NVIDIA/cuda-quantum/compare/0.15.0...0.16.0