KalshiClosedCheck availabilityKYC requiredVerified resolution dataWell calibratedThis market resolved: Before Sep 18, 2026 (99%)
Resolved: Sep 7, 2026, 3:43 PM
Provider-verified resolution
CoinRithm has provider-backed winner and settlement-time evidence for this market.
Data quality warningStale data
Data as of Sep 7, 2026, 3:05 PM UTC · policy pm-quality-3
Probability Timeline
Before Aug 14, 20262%
Before Oct 16, 202699%
Before Sep 18, 202699%
Before Dec 1, 202699%
Before Nov 1, 202699%
Before Sep 18, 2026Won
Before Oct 1, 2026Lost
Before Oct 16, 2026Lost
Before Nov 1, 2026Lost
Before Dec 1, 2026Lost
Before Aug 14, 2026Lost
Before Aug 21, 2026Lost
Before Sep 1, 2026Lost
Related Markets
Related questions for context, not a comparison of equivalent contracts.
Rules
Archived event-level rules. The exact terms for each outcome are no longer available.
If OpenAI releases Astra the named outcome, then the market resolves to Yes.
Settles via:OpenAI
- Release must be to the public, outside of a closed beta, though limiting it to a high-cost subscription tier is acceptable.
- On August 1, 2026, OpenAI publicly disclosed "Astra," an internal model described by OpenAI as its next major frontier system, citing results including the resolution of ten previously open mathematics problems - https://openai.com/index/ten-advances-in-mathematics/
- Qualifying Release: The Payout Criterion is satisfied if OpenAI releases a model that either (a) is publicly named "Astra" (including numbered, lettered, or suffixed variants such as "Astra 1," "Astra 6," or "GPT-6 Astra"), or (b) is confirmed by OpenAI, or by a consensus of credible reporting, to share the same underlying model — including the same base architecture, weights, or training run — as the model referenced in OpenAI's August 1, 2026 disclosure, regardless of the commercial name, safety configuration, or access tier under which it is ultimately released.
- A separately-branded release that is derived from or built on the same underlying model as the disclosed Astra model satisfies this clause even if OpenAI markets it as a distinct product (e.g., a safety-restricted variant or a differently-priced tier), provided the shared underlying model is confirmed by OpenAI or by a consensus of credible reporting.