Skip to main content
    All AI News
    OpenAI News (firm-scan)Thursday, September 24, 2026 3 min read
    AI

    Introducing GPT-6 Sol and GPT-6 Luna

    OpenAI has launched GPT-6 Sol and GPT-6 Luna, two cost-optimized models that extend the GPT-6 family below the flagship GPT-6 Astra. Both models are priced 50% below their GPT-5.6 predecessors, with Sol at $2 input/$10 output and Luna at…

    Key takeaways
    • 01On key benchmarks, GPT-6 Sol outperforms Claude Opus 5 on business automation tasks at roughly 9% of the cost, and matches Claude Fable 5.1 on software engineering at approximately 80% lower cost per task.
    • 02GPT-6 Luna achieves comparable coding performance to Claude Opus 5 and Fable 5 at medium effort while costing 93-96% less.
    • 03OpenAI also improved prompt caching to deliver 90% discounts on cached input tokens, further reducing costs for agents and long-context applications.
    In brief · from openai.com

    OpenAI has launched GPT-6 Sol and GPT-6 Luna, two cost-optimized models that extend the GPT-6 family below the flagship GPT-6 Astra. Both models are priced 50% below their GPT-5.6 predecessors, with Sol at $2 input/$10 output and Luna at $0.10 input/$0.50 output per million tokens. On key benchmarks, GPT-6 Sol outperforms Claude Opus 5 on business automation tasks at roughly 9% of the cost, and matches Claude Fable 5.1 on software engineering at approximately 80% lower cost per task.

    Read the full article at openai.com

    OpenAI has launched GPT-6 Sol and GPT-6 Luna, two cost-optimized models that extend the GPT-6 family below the flagship GPT-6 Astra. Both models are priced 50% below their GPT-5.6 predecessors, with Sol at $2 input/$10 output and Luna at $0.10 input/$0.50 output per million tokens. On key benchmarks, GPT-6 Sol outperforms Claude Opus 5 on business automation tasks at roughly 9% of the cost, and matches Claude Fable 5.1 on software engineering at approximately 80% lower cost per task. GPT-6 Luna achieves comparable coding performance to Claude Opus 5 and Fable 5 at medium effort while costing 93-96% less. OpenAI also improved prompt caching to deliver 90% discounts on cached input tokens, further reducing costs for agents and long-context applications.

    Don't miss tomorrow's

    The Daily Pulse in your inbox each morning — sourced and linked.

    How often
    Keep going — across the app