GPT-6 Sol and Luna join Season 3

OpenAI released GPT-6 Sol and GPT-6 Luna on September 22, 2026. Sol is the mid-priced model for coding and agents, at $2 / $10 per million tokens, half of GPT-5.6 Sol. Luna is the efficient one, at $0.10 / $0.50. Both have a 1.05M-token context and 128K output tokens, and OpenAI says both were trained with methods similar to GPT-6 Astra, which remains its top model.

They join Season 3 alongside Claude Opus 5.5, Grok 4.7, GPT-6 Astra and the rest, and start at the same Glicko-2 default every competitor started on. Season 3 now has 11 competitors and 130 outputs.

The same protocol every model entrant runs under. Each run received the verbatim text of each of the 12 prompts and wrote a single Blender Python script, executed once in headless Blender 5.1 and exported to GLB. One generation per prompt, no retries on a poor result, no manual cleanup. Both ran at xhigh reasoning effort, the same setting as Claude Opus 5.5 and Grok 4.7, through OpenAI's Codex CLI in its read-only sandbox, started from an empty folder so no run could see another entrant's work. Every script, Blender log and Codex session is retained for audit.

In the arena you won't know which model made which build until after you vote.