I want to make notes about the announcements done by main companies in this area, so I can review them in future.<\/p>\n\n\n\n
The table below covers 10 listed quantum companies and the technical milestones they have publicly committed to for roughly Q4 2026 through 2029.<\/p>\n\n\n\n#<\/th> Company (ticker)<\/th> Late 2026<\/th> 2027<\/th> 2028<\/th> 2029+<\/th><\/tr><\/thead> 1<\/td> IonQ<\/strong> (IONQ)<\/td> Combined IonQ and SkyWater 2026 revenue guidance of $450\u2013460M. SkyWater is already producing “Superion” chips.<\/td> First 256-qubit Superion customer deployments (early 2027).<\/td> 200,000-qubit QPUs (8,000 logical qubits) start functional testing.<\/td> Roadmap to about 2M qubits by 2030.<\/td><\/tr> 2<\/td> Rigetti<\/strong> (RGTI)<\/td> 108-qubit Cepheus-1 is generally available. The published roadmap targeted a 150+ qubit system at 99.7% fidelity by end of 2026, and the company says it will update the roadmap later this year.<\/td> The 2025 roadmap targeted 1,000+ qubits at 99.8% fidelity by end of 2027. The Q2 2026 release describes about 1,000 qubits at about 99.9% fidelity and gates under 50 ns over roughly three years. HPE and Pittsburgh Supercomputing Center delivery is expected in 2027.<\/td> Continued scaling toward the 1,000-qubit target.<\/td> Up to $100M UK investment for system deployments.<\/td><\/tr> 3<\/td> D-Wave<\/strong> (QBTS)<\/td> 17-physical-qubit gate-model system (logical error rate 2x lower than physical).<\/td> 49 physical qubits (20x error reduction).<\/td> 181 physical qubits (2,000x error reduction).<\/td> 20,000-qubit Advantage3 annealer (2029). 10 logical qubits (2030). 100,000-qubit annealer (2031). 100 logical qubits (2032).<\/td><\/tr> 4<\/td> IBM<\/strong> (IBM)<\/td> Kookaburra: first module combining qLDPC memory with a logical processing unit. Nighthawk runs 7,500-gate circuits on up to three 120-qubit modules.<\/td> Cockatoo: entanglement between modules. Nighthawk improves to 15,000 gates on up to 1,080 qubits.<\/td> Magic-state injection across multiple modules.<\/td> Starling (2029): 200 logical qubits, 100M gates, in Poughkeepsie. Blue Jay (2,000 logical qubits, 1B ops) after 2033.<\/td><\/tr> 5<\/td> Quantinuum<\/strong> (QNT)<\/td> Helios (98 physical, 48 logical qubits) is being deployed.<\/td> Sol: 192 physical and 100 logical qubits, 10\u207b\u2075 logical error rate, first 2D QCCD trap chip.<\/td> Continued de-risking toward Apollo.<\/td> Apollo (2029): thousands of physical and hundreds of logical qubits.<\/td><\/tr> 6<\/td> Infleqtion<\/strong> (INFQ)<\/td> 30 entangled logical qubits on 80 physical qubits was announced Sept 24, 2026.<\/td> Illinois neutral-atom system, modular to more than 50 logical qubits.<\/td> 100 logical qubits.<\/td> 1,000+ logical qubits by 2030.<\/td><\/tr> 7<\/td> Xanadu<\/strong> (XNDU)<\/td> Qubit Factory construction begins (2026\u201327). Photon-loss indicator at 24.1x.<\/td> Qubit Factory build continues.<\/td> Fault tolerance (2028\u201329).<\/td> Up to 200 logical qubits (2029), 500 (2030), 1,000+ (2031). Quantum data center in 2029\u201330.<\/td><\/tr> 8<\/td> IQM<\/strong> (IQMX)<\/td> 150-qubit Halocene commercially available by end of 2026. It targets 99.7% fidelity and up to 5 logical qubits. Munich system upgraded to 150 qubits.<\/td> LUMI-IQ starts with a 150-qubit Halocene H4. 300-qubit system to VTT.<\/td> Intermediate LUMI-IQ upgrade.<\/td> Halocene H5 (2029): up to 9 logical qubits with lattice surgery and T-gates. Fault tolerance by 2030.<\/td><\/tr> 9<\/td> Microsoft<\/strong> (MSFT)<\/td> Majorana 2 was announced in June 2026. The published device roadmap moves from single-qubit devices to two-qubit braiding devices and then a 4\u00d72 tetron array. That array is meant to show error detection on two logical qubits.<\/td> No dated milestone.
Microsoft says it intends to build a fault-tolerant prototype topological system “in years, not decades” under DARPA’s US2QC program.<\/td>No dated milestone.<\/td> The roadmap points to a 27\u00d713 tetron array for quantum error correction, with no date. Majorana 1 is described as designed to scale to a million qubits on one chip.<\/td><\/tr> 10<\/td> Alphabet \/ Google Quantum AI<\/strong> (GOOGL)<\/td> Willow Early Access Program is moving quantum hardware to outside research partners. The stated next roadmap milestone is a long-lived logical qubit, with no date.<\/td> No dated technical target.<\/td> No dated technical target.<\/td> Google has framed a useful, error-corrected quantum computer around 2029, with about a million physical qubits as the eventual scale. Willow currently has 105 physical qubits.<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n