Will Rigetti report a median two-qubit gate fidelity of at least 99.5% on its 108-qubit Cepheus-1-108Q system before 31 December 2026?
deadline 2026-12-31
confidence medium
tier 4 evidence
39%Asturna estimate
Resolution criterion
Yes if Rigetti publicly reports (press release, SEC filing, earnings release or investor presentation) a median two-qubit gate fidelity of at least 99.5% measured on Cepheus-1-108Q (or its 108-qubit system), on or before 31 December 2026. Best-pair, single-chiplet or other non-median figures do not count.
Verified starting state
Rigetti announced general availability of Cepheus-1-108Q in April 2026 with a 99.1% median two-qubit fidelity. On its Q2 call (7 August 2026) the median was still about 99.1%; management expects about 99.5% 'later in 2026' and is testing carefully before a cloud rollout. The 108-qubit system had already been delayed once (January 2026).
What pushes it up
- Rigetti executive management explicitly stated on the August 2026 Q2 call that they expect median two-qubit gate fidelity to reach approximately 99.5% later in 2026 [tier 1].
- Cepheus-1-108Q has been actively deployed across cloud platforms like AWS Braket and qBraid since mid-2026, providing continuous operational testing and calibration data [tier 4].
What pushes it down
- Increasing median two-qubit fidelity from 99.1% to 99.5% across an entire 108-qubit chip requires cutting the error rate by over 40%, which is a difficult technical hurdle [tier 1].
- Fidelity remained flat at approximately 99.1% between the April launch and the August Q2 call, showing lack of rapid progress over that four-month span [tier 1].
- With only two and a half months remaining before the 31 December 2026 deadline, any delay in testing or calibration will push resolution into 2027, echoing previous timeline slips on this architecture [tier 1].
What would move this number most
Whether calibration and control optimization alone can achieve 99.5% median fidelity across all 108 qubits in time for Q3 earnings or an end-of-year update without requiring a hardware revision.
How this number is built
| Deadline | 2026-12-31 |
| Base rate | 35% |
| Best evidence | tier 4 · 2 article(s) used |
| Independent runs | 38 · 40 (spread 2 pts) |
| Confidence | medium |
| Evidence cap | not triggered |
| Last revised | 2026-10-15 15:02 |
| Model | gemini-3.7-flash |
Sources consulted
Probability history
2026-10-15 15:02 39% 7 article(s) on file · gemini-3.7-flash
- What pushes it up. Rigetti executive management explicitly stated on the August 2026 Q2 call that they expect median two-qubit gate fidelity to reach approximately 99.5% later in 2026 [tier 1].
- What pushes it up. Cepheus-1-108Q has been actively deployed across cloud platforms like AWS Braket and qBraid since mid-2026, providing continuous operational testing and calibration data [tier 4].
- What pushes it down. Increasing median two-qubit fidelity from 99.1% to 99.5% across an entire 108-qubit chip requires cutting the error rate by over 40%, which is a difficult technical hurdle [tier 1].
- What pushes it down. Fidelity remained flat at approximately 99.1% between the April launch and the August Q2 call, showing lack of rapid progress over that four-month span [tier 1].
- What pushes it down. With only two and a half months remaining before the 31 December 2026 deadline, any delay in testing or calibration will push resolution into 2027, echoing previous timeline slips on this architecture [tier 1].
- What would move this number most. Whether calibration and control optimization alone can achieve 99.5% median fidelity across all 108 qubits in time for Q3 earnings or an end-of-year update without requiring a hardware revision.
2026-10-14 14:17 18% 6 article(s) on file · openai/gpt-oss-20b (Groq)
- What pushes it up. Management expects 99.5% median two‑qubit fidelity later in 2026, indicating a planned improvement and a likely report before the deadline [tier 1]
- What pushes it down. Median fidelity remains at ~99.1% as of August 2026 with no reported improvement, and the 108‑qubit system has already experienced a delay, suggesting technical challenges that could postpone or prevent the target [tier 1]
- What would move this number most. Whether Rigetti will actually achieve and publicly report ≥99.5% median two‑qubit fidelity on the 108‑qubit system before 31 Dec 2026