GPT-6 Sol and GPT-6 Luna Now Available in AI Agent and GPT Actions

OpenAI’s newest models, GPT-6 Sol and GPT-6 Luna, are now available in the AI Agent action and the GPT action in Workflows. They replace the GPT-5.6 models, with more accurate outputs, cleaner responses, and pricing that is 50 percent lower.

GPT-6 Sol: The model for complex work. It is built for multi-step agent tasks, business workflows that span multiple apps, and decisions that need careful reasoning. GPT-6 Sol makes roughly half as many factual mistakes as GPT-5.6 Sol.

GPT-6 Luna: The fast, efficient model for high-volume tasks such as classification, summaries, tagging, and quick replies. At higher effort settings, GPT-6 Luna matches GPT-5.6 Sol on factual accuracy at a fraction of the cost.

Model Selection: Both models appear in the Model dropdown of the AI Agent action and the GPT action.

More reliable automations: Fewer factual mistakes mean fewer bad outputs flowing into downstream steps, such as a wrong tag, a misrouted contact, or an inaccurate reply.

Better results on real business workflows: On OpenAI’s benchmarks for end-to-end workflows across sales, marketing, support, and operations, both models outperform their GPT-5.6 predecessors. GPT-6 Luna delivers better results at up to 58 percent lower cost per task.

Cleaner outputs: Responses are clearer and tighter, with less jargon and no loss of substance, so generated content needs less cleanup before it reaches customers.

AI at lower cost: Both models cost half of what GPT-5.6 Sol and Luna did, which makes it practical to run AI steps across more workflows and at higher volumes.