Model evaluation roster

This is the planned Season 1 roster, not a ranking or a claim that every model has completed evaluation. Model names and configurations reflect the October 1 pilot snapshot. See the pilot results and limitations for completed attempts.

• Claude Fable 5.1Anthropic · awaiting comparable results
• GPT-6 AstraOpenAI · awaiting comparable results
• Claude Opus 5.5Anthropic · awaiting comparable results
• GPT-6.1 SolOpenAI · awaiting comparable results
• Gemini 3.1 ProGoogle · awaiting comparable results
• Claude Sonnet 5.5Anthropic · awaiting comparable results
• GPT-5.3 CodexOpenAI · awaiting comparable results
• Grok 4.7xAI · awaiting comparable results
• Gemini 3.8 FlashGoogle · awaiting comparable results
• Muse Spark 1.3Meta · awaiting comparable results
• Kimi K2.7 CodeMoonshot · awaiting comparable results
• Claude Haiku 4.5Anthropic · awaiting comparable results
• GPT-6 LunaOpenAI · awaiting comparable results

Choosing a model now? Start with the PCB evaluation checklist rather than assuming one model wins every task.