GPT-6 Release Date Odds: Pricing OpenAI's Ship Cadence

The GPT-6 release date is not a date yet, but it already has a price. On Kalshi's "When will OpenAI release GPT-6?" event, the contract for a GPT-6-or-greater release before November 1, 2026 trades at a 58-cent bid against a 60-cent ask, roughly a 59% implied probability, while the rung for before September 1 sits down at 10 to 15 cents. That curve is the crowd's answer sheet, and its shape makes a specific claim: the next numbered flagship lands this fall, at nearly double the pace OpenAI's own release history supports. This page reads that gap the way a sims player reads a projection against an ownership column: start from the base rate, update on the news, and check the fine print before trusting the number. The fine print matters more than usual here, because this board is quietly two bets in one.

The Quick Answer

Kalshi prices a public GPT-6-or-greater release before November 1, 2026 at 58 to 60 cents, about a 59% implied chance, with before September 1 near 13% and before 2027 last trading at 80 cents on a very wide book. OpenAI's last two flagship cycles say numbered releases arrive roughly every two and a half years, which would put GPT-6 in 2027 or later, so the market is pricing news, not cadence. The full timing ladder, the cadence math, and the naming clause that splits this bet in two are below.

The Market At A Glance

Those are the mechanics. The curve they produce is the interesting part, because it is anything but smooth.

The Timing Ladder Has One Fat Month

Live Kalshi order book, August 12, 2026. These are cumulative thresholds, so probabilities climb with the date; never add the rungs together.

Because each rung is cumulative, the information lives in the differences, exactly like reading a sims output that reports finish-by thresholds. Subtract September's roughly 13% from where October trades and the crowd is stuffing roughly 32% to 38% of its probability into September alone, depending on whether you read October's wide book at its 45-cent last trade (a 45% read) or its 51-cent midpoint (a 51% read). October itself adds single digits to the low teens, and the rest of the year adds maybe 20 more, though that last figure leans on a stale print from the widest book on the board (more on that below). Slice the range however you like; one month still carries the forecast, which means the market is not really pricing a cadence at all. It is pricing an event on a calendar. Hold that thought, because the base rate says something very different.

The Base Rate: What OpenAI's Cadence Actually Supports

Numbered flagships have never come fast. GPT-2 arrived in February 2019, GPT-3 in June 2020, GPT-4 in March 2023, and GPT-5 in August 2025. That is 16 months, then 33, then 29 between major numbers, an average of about two and a half years across the last two cycles. A GPT-6 before November 1, 2026 would land roughly 15 months after GPT-5, half the recent gap. A forecast built on cadence alone would price the November rung far below the 58 to 60 cents on the screen today.

The counterargument is that the treadmill underneath the numbers has sped up. OpenAI shipped GPT-5.5 in April 2026 and then GPT-5.6 in July 2026, a preview on June 26 and general availability on July 9 in three tiers named Luna, Terra, and Sol. Two ships three months apart is real velocity. But notice what neither ship did: neither one carried a new number. The point releases prove OpenAI can ship quickly and simultaneously prove it has been choosing not to spend the "6" on those ships.

So why is the market at 59% anyway? Because since early August, it has stopped pricing cadence and started pricing one announcement.

The Update: Astra Moved The Curve

In early August 2026, OpenAI introduced Astra, and it did so in the strangest way a frontier lab has ever teed up a product: with solved open problems in mathematics rather than a launch. There is no release date, no pricing, and no model card. Subsequent reporting points to a planned launch by November 2026, and that reported window lines up almost perfectly with the board above: a fat September, a November rung near 60 cents, and a before-2027 print in the 80s. The crowd is doing textbook forecasting, holding a slow prior and then updating hard on verified news. The base rate is not being ignored; the market is saying the base rate went stale the moment the company publicly queued its next release.

This is the same two-column discipline a DFS player runs every slate: the projection is the base rate, the news is the update, and the number you actually use is neither one alone. It is why process beats vibes in both arenas. The MLB DFS projections, ownership and stacks in the DataHub exist to run exactly this arithmetic on ballplayers instead of language models, and you can try the Sims free to watch base rates and news get blended thousands of times per slate.

The Naming Fine Print Splits The Bet

Here is the second bet I promised in the opening. The market's rules resolve YES only if OpenAI releases "a model called GPT-6 or greater" before the date, with the release public and outside a closed beta (a high-cost subscription tier still counts). Astra is not confirmed to be GPT-6. OpenAI has not said whether it ships under that name or as yet another GPT-5 point release, and the two most recent ships, 5.5 and 5.6, both took the point-release exit. By the rules text, an on-time Astra that launches under the Astra name alone would not obviously settle these contracts YES.

That means the roughly 40 cents of NO on the November rung is not all timing risk. Part of it is naming risk, and NO holders win either way: Astra slips past November, or Astra arrives on schedule wearing a different badge. YES needs both the calendar and the christening to cooperate. In leverage terms, the November YES is a narrow-path play priced like chalk, and recognizing a narrow path behind a confident price is the most transferable skill on this page. Read the rules on the market page yourself before forming an opinion; resolution language is the box score of this arena.

A Worked Example: From Cents To Probability

An event contract pays $1 if it settles YES and $0 if it settles NO, so a 60-cent price is the crowd pricing roughly a 60% chance, the same way 13 cents is about 13%. Fees shift the effective number. Kalshi's published trading fee formula is 0.07 times contracts times price times one minus price: fill 100 November contracts at 60 cents and the fee is 0.07 x 100 x 0.60 x 0.40, about $1.68 on a $60 position, roughly 2.8% of the stake. Your true break-even sits closer to 62% than 60%. Fees bite hardest near 50 cents and fade toward the extremes, so the mid-curve rungs, exactly where this board's action is, carry the biggest haircut.

The before-2027 rung teaches the other reading habit. Its last trade says 80 cents, but the live book is a 40-cent bid against an 82-cent ask, a 42-cent-wide window. On a book that wide, the last print is a stale opinion, not a price, the same way Tuesday's ownership projection is not Friday's slate. Read the range, treat the midpoint as a guess, and if you ever act on a market this thin, size in fractions of a bankroll unit, because the spread taxes you at the door and again at the exit.

FAQ

When Will GPT-6 Be Released?

OpenAI has not announced a GPT-6 release date. As of August 12, 2026, Kalshi's market prices a GPT-6-or-greater public release before November 1, 2026 at 58 to 60 cents, roughly 59%, and before September 1 at 10 to 15 cents. The rungs for before July 1 and before August 1, 2026 have already settled NO.

Does Astra Count As GPT-6 For This Market?

Only if the name cooperates. The contracts resolve YES for a public release of a model called GPT-6 or greater, and OpenAI has not said whether Astra ships as GPT-6 or as another GPT-5 point release. An Astra launch under a different name would not, on the rules text alone, settle these contracts YES. Check the resolution rules on Kalshi's market page before treating the November price as a pure timing number.

What Does OpenAI's Release History Say About The Timing?

Numbered flagships shipped in February 2019 (GPT-2), June 2020 (GPT-3), March 2023 (GPT-4), and August 2025 (GPT-5), gaps of 16 to 33 months. The point releases move faster, with GPT-5.5 in April 2026 and GPT-5.6 in July 2026. A cadence-only forecast lands GPT-6 well past 2026, which is exactly the prior the market abandoned when Astra was introduced in early August.

What Is The 40 Cents Of NO On November Paying For?

Two different risks at once. Astra could slip past November 1, or it could arrive on schedule under a name that is not GPT-6 or greater, and the contract's rules pay NO in both branches. That dual coverage is why the November rung is harder to read than a pure timing market. A standing reminder while you read it: these prices are a live crowd estimate, not a forecast of fact and not financial advice, event contracts are for adults 18 and older where legal with availability varying by state as of August 2026, and one OpenAI blog post can repaint the entire curve in an afternoon.

The Bottom Line

The GPT-6 release date market is the cleanest forecasting exercise on the board right now: a two-and-a-half-year base rate, one early-August announcement that cut the market's patience in half, and a naming clause that turns a confident 59% into a narrow-path bet with two ways to lose. The fat September month is the tell; this is an event market wearing a cadence market's clothes. This page runs on live market prices and public facts; our price-blind AI model panel scores this board on the page's next refresh, with every panel seat graded against real market settlements as they resolve. For more boards read this way, start at our sports prediction markets hub; the cumulative-curve habit transfers directly to the MLB division race odds, and the narrow-path chalk read repeats on the next-team boards for Bronny James and Tyreek Hill. The award fields tell the attention-versus-probability side of the story, in the NBA MVP odds, the Heisman Trophy odds, and the MLB Rookie of the Year odds. Prices move; the reading habit is the durable part.