Breaking

OpenAI GPT model debut odds plunge 18pp to 12% as whale flow diverges

Whales added $4K to YES contracts even as Polymarket prices the GPT model debut significantly lower.

The probability that the next GPT model released by OpenAI will debut at a score of at least 1490 by December 31, 2026, fell sharply by 18.0 percentage points over the last 24 hours on Polymarket, dropping from an estimated 30.0% to 12.0%.

Despite this steep decline in the YES price, whale activity diverged from the market’s direction. Data shows a net $4K inflow into YES contracts, with $8K in whale buy volume against $4K in sells, spread across 27 unique whales. This suggests that large traders increased their exposure to the YES outcome even as the broader market repriced it substantially lower.

The 24-hour volume on this market reached $7K, while the lifetime volume stands at $58K among 205 unique traders. Notably, the Polymarket on-chain mid-price for YES contracts remains at 45.5%, much higher than the current Polymarket feed price of 12.0%, indicating a significant discrepancy between on-chain data and the platform’s quoted price.

The contrasting signals from price and whale activity underscore the uncertainty surrounding the GPT model’s performance expectations through the end of 2026.

Market Will the next GPT model released by OpenAI debut at a score of at least 1490 by December 31, 2026?
Market ID 2668022
24h price change +18.0 pp
YES now (PM Breaking) 12.0%
YES ~24h ago (est.) 30.0%
YES (Polydata overview) 45.5%
Whale net flow (24h) $4K
Whale buy / sell (24h) $8K / $4K
Unique whales (24h) 27
Volume 24h (PM) $7K
Unique traders (Polydata) 205

Source: Polydata API v3 · /whales/flow + Polymarket Breaking · snapshot 2026-07-09. Data: Polydata API v3. On-chain figures are public. Realized PnL is computed over resolved markets only and excludes open positions, so it is conservative versus the Polymarket UI. This is not investment advice.

Read next

archive →