Luna Included in GPT-5.6 Model Family, Controversy Over Sub-Agent Control Method

OpenAI's GPT-5.6 model family includes Luna, and the multi-agent feature is labeled as 'multi-agent beta.' In a statement released on July 9, OpenAI presented Luna as a cost-effective model. Luna is used for high-frequency tasks and is structured as a sub-model within the GPT-5.6 product line rather than a separate model. The issue at hand is not the existence of Luna but rather the method of invoking and controlling sub-agents. OpenAI's developer documentation explains that 'multi-agent beta' refers to the capability of coordinating multiple sub-agents in parallel and merging their results. CryptoBriefing reported the addition of Luna support, but official documents suggest that 'multi-agent beta' is more appropriate. OpenAI announced on July 30 that it reduced Luna's price by 80% and Terra's price by 20%. Developers are focusing more on control issues than on feature support, with claims raised in GitHub issues that custom agent selection has been blocked. Multi-agent systems involve a single AI system delegating tasks to multiple sub-agents and integrating the results. Current public documents do not definitively state that 'multi-agent v2 has newly supported Luna.'

This content is provided for general informational purposes only and doesn't constitute financial, investment, legal, or tax advice. Any events, rewards, online promotions, or related information mentioned herein should not be considered a recommendation, solicitation, or invitation to purchase, sell, trade, or otherwise deal in any crypto assets. Crypto assets are highly volatile and may result in loss. The availability of WEEX services, products, and related events may vary by region. You are responsible for ensuring that your participation is in accordance with applicable local laws and regulations.