Resolves to the next GPT model released regardless of number. Only the “stock” GPT counts, so no mini, nano, pro, codex, etc

• Update 2026-06-26 (PST) (AI summary of creator comment): The model "Sol" will count as the relevant GPT model for this market, as its pricing is comparable to GPT-5.5.

• Update 2026-07-09 (PST) (AI summary of creator comment): The creator is waiting for Epoch to publish results before resolving. Based on the OpenAI page, GPT-5.6 (Sol) appears to score well below Claude Fable on FrontierMath Tier 4.