Key Highlights

• OpenAI has introduced the GPT 5.6 family with three new models: Sol, Terra, and Luna.
• The company says GPT 5.6 delivers stronger coding, cybersecurity, and enterprise performance while using fewer tokens.
• OpenAI also announced ChatGPT Work, a workplace-focused AI assistant for business teams.
• The new models are now available across ChatGPT, Codex, and the OpenAI API with tiered pricing.

OpenAI has officially launched GPT-5.6, introducing a new family of AI models designed to improve coding, cybersecurity, enterprise productivity, and scientific research. The GPT-5.6 lineup includes Sol, Terra, and Luna, each targeting different performance and pricing needs while expanding the company’s enterprise AI portfolio.

The announcement marks OpenAI’s latest move in the increasingly competitive artificial intelligence market. Alongside the new models, the company also introduced ChatGPT Work, a workplace assistant built to help teams create documents, spreadsheets, and presentations across desktop, web, and mobile.

What is OpenAI GPT-5.6?

The GPT-5.6 family represents OpenAI’s newest generation of large language models. Instead of releasing a single flagship model, the company has introduced three variants that balance performance and cost.

Sol serves as the flagship model. OpenAI positions it as its most capable option for software development, enterprise workflows, and advanced reasoning tasks.

Terra targets organizations that need strong AI performance at a lower cost. It offers many of the same capabilities while reducing compute requirements.

Luna is the entry-level model. It focuses on affordability while still delivering competitive performance for everyday AI workloads.

This tiered strategy allows businesses and developers to select a model based on their budget and performance requirements instead of relying on a one-size-fits-all approach.

Why is OpenAI focusing on coding and cybersecurity?

One of the biggest announcements surrounding GPT-5.6 is its emphasis on coding and cybersecurity.

According to OpenAI, GPT-5.6 is its strongest cybersecurity model so far. The company says the models can support defensive security tasks such as threat modeling, reviewing code for vulnerabilities, recommending security patches, and assisting with blue-team exercises. These simulations help organizations identify weaknesses before attackers can exploit them.

CEO Sam Altman recently stated that Sol is 54% more token-efficient for AI coding tasks compared to previous models. That means developers can complete complex coding workflows while using fewer tokens, potentially lowering operating costs and improving response speed.

The cybersecurity capabilities have also drawn attention because advanced AI security tools continue to raise discussions around responsible deployment. OpenAI says the models are intended to strengthen defensive security rather than offensive cyber operations.

How does GPT-5.6 compare with rival AI models?

OpenAI is making direct comparisons with competitors, particularly Anthropic.

The company cited results from the Artificial Analysis Coding Agent Index, claiming that Sol achieves a score of 80, outperforming Anthropic’s recently introduced Fable 5. OpenAI also says Sol reaches those results while using less than half the output tokens, completing tasks in less than half the time, and costing roughly one-third less.

The comparisons extend beyond the flagship model. OpenAI claims Terra performs slightly above Fable 5, while Luna outperforms Anthropic’s Opus 4.8 in benchmark testing.

These claims highlight how benchmark performance has become a major battleground as AI companies compete for enterprise customers. Faster responses, lower costs, and improved coding performance have become key selling points for businesses deploying AI at scale.

What is ChatGPT Work?

Alongside GPT-5.6, OpenAI introduced ChatGPT Work, a productivity-focused assistant built specifically for enterprise users.

The tool works across desktop, mobile, and web platforms. It helps employees draft reports, prepare spreadsheets, create presentations, summarize information, and manage routine office tasks.

Rather than replacing existing productivity software, ChatGPT Work is designed to assist teams throughout their daily workflows. This reflects the broader trend of AI becoming an integrated workplace companion instead of a standalone chatbot.

The launch also signals OpenAI’s growing focus on enterprise software, where long-term subscriptions and organizational deployments continue to drive revenue growth.

Pricing and availability

OpenAI has made the GPT-5.6 family available immediately through ChatGPT, Codex, and the OpenAI API.

The pricing structure reflects the three performance tiers. Sol costs $5 per million input tokens and $30 per million output tokens. Terra is priced at $2.50 per million input tokens and $15 per million output tokens. Luna offers the lowest entry point at $1 per million input tokens and $6 per million output tokens.

This pricing strategy gives developers flexibility while making advanced AI models accessible to organizations with different budgets.

Why this launch matters

The release of GPT-5.6 arrives during one of the most competitive periods in the AI industry. Several leading companies have recently introduced new models, intensifying competition across enterprise software, coding assistants, and productivity platforms.

Instead of focusing solely on chatbot improvements, OpenAI is positioning GPT-5.6 as a practical business platform. Better coding performance, stronger cybersecurity capabilities, enterprise productivity tools, and lower operating costs all reflect that strategy.

As organizations continue integrating AI into everyday workflows, efficiency and pricing are becoming just as important as raw model performance. With GPT-5.6, OpenAI is clearly targeting that shift while strengthening its position in the enterprise AI market. Whether these improvements translate into broader enterprise adoption will become clearer in the months ahead, but GPT-5.6 represents one of OpenAI’s most significant product launches focused on business users to date.