Rigetti announced an expanded collaboration with HPE and the Pittsburgh Supercomputing Center on July 27, 2026, to build TangleLab, a hybrid quantum-classical supercomputing testbed. The project is funded by a $5 million National Science Foundation grant, pairs a 9-qubit Rigetti Novera system with a classical HPC system, and construction begins September 1, 2026, at PSC's new data center, targeting full operations in 2027.
A small qubit count, on purpose
Nine qubits is a modest number next to the systems this site usually covers, and that's the point of a testbed rather than a production system. TangleLab exists to work out the integration problem, how a real quantum processor and a real classical supercomputer talk to each other under production HPC conditions, not to demonstrate a qubit-count milestone. Our ranking of the top quantum computing companies already covers where Rigetti sits on pure hardware metrics: behind IBM, Google, IonQ, and Quantinuum on fidelity and commercial traction, with its strongest asset being a solid cash position. A federally funded integration testbed is a different kind of contribution, orthogonal to that ranking rather than a challenge to it.
A different integration model than NVIDIA's
Our NVQLink page covers NVIDIA's approach to the same general problem, connecting GPU-accelerated classical compute to QPU control electronics with sub-4-microsecond latency, aimed at real-time error correction decoding. TangleLab is a different kind of integration project: an NSF-funded, academically hosted testbed built around one specific vendor's hardware (Rigetti's Novera) rather than an open, vendor-agnostic architecture spanning many QPU builders. Both are real answers to "how does a quantum processor plug into HPC infrastructure," built for different purposes, one a broad industry standard, one a focused research testbed.
What's confirmed and what's still ahead
Confirmed: the $5 million NSF grant, the 9-qubit Novera hardware, the September 1 construction start, and the 2027 full-operations target. Not yet known: what specific research workloads TangleLab will run first, or what integration lessons it produces that might generalize beyond this one testbed. That's the part worth watching once construction finishes.
What to watch next
Whether TangleLab publishes real integration benchmarks (latency, throughput between the classical and quantum sides) once operational in 2027, the same kind of number this site looks for in any hybrid quantum-classical claim. Our hardware overview tracks how testbeds like this compare with other paths to real QPU access.