OpenAI has just dropped a bombshell announcement that goes far beyond a typical model upgrade. Introducing GPT-5.6, featuring new models Sol, Terra, and Luna, alongside the powerful new productivity tool, ChatGPT Work. These releases signal a clear and aggressive move by OpenAI to challenge its competitor, Anthropic, head-on in the fiercely competitive AI landscape.
For months, the AI world has buzzed with excitement around Anthropic’s offerings like Fable 5 and Mythos 5. It appears OpenAI isn’t just participating in the conversation; they’re determined to lead it. With these new tools, OpenAI aims to outshine rivals on critical fronts: price, speed, and overall productivity for users and enterprises alike.
OpenAI’s Next-Gen Models: GPT-5.6 Unleashed
The arrival of GPT-5.6 has been highly anticipated, especially after its release was temporarily delayed by government intervention last month. This new iteration comes in three distinct flavors—Sol, Terra, and Luna—each designed to push the boundaries of AI performance. OpenAI’s briefings highlighted specific benchmarks, directly targeting Anthropic’s performance metrics.
When it comes to handling complex, multi-step professional workflows across dozens of fields, GPT-5.6 Sol scored an impressive 53.6 on the Agents’ Last Exam benchmark. This result significantly surpassed Claude Fable 5 by a considerable 13.1 points, demonstrating Sol’s superior capability in intricate tasks. OpenAI isn’t just claiming leadership; they’re backing it up with hard data.
The new models also show significant strides in general intelligence and efficiency. On the Artificial Analysis Intelligence Index, GPT-5.6 Sol came within one point of Fable 5 while completing tasks in an astonishing 61% less time. Furthermore, in coding proficiency, Sol achieved an 80 on the Artificial Analysis Coding Agent Index, beating Fable 5 by 2.8 points.
Beyond raw power, efficiency and cost-effectiveness are major selling points for GPT-5.6. OpenAI reports that Sol is considerably more efficient and therefore less costly than Fable, particularly at medium reasoning levels. At this complexity, Sol reportedly beat Fable 5 by 11.4 points while costing approximately one-quarter of the estimated price.
The improvements aren’t just against competitors but also within OpenAI’s own lineage. GPT-5.6 is faster and more efficient than its predecessor, GPT-5.5. Itamar Friedman, co-founder and CEO of AI code review tool Qodo, noted that GPT-5.6 was the strongest model they evaluated on agentic code-review tests, using roughly three times fewer tokens per pull request and delivering about two times lower median latency compared to GPT-5.5.
Following recent industry discussions around AI safety, OpenAI emphasized that GPT-5.6 includes their “most robust safeguards to date.” These protections integrate advanced model guardrails, real-time checks, continuous monitoring, and access calibrated to trust and risk levels. This commitment to safety aims to address concerns and build greater user confidence.
Meet ChatGPT Work: Your New AI Productivity Partner
Beyond the core models, OpenAI also unveiled ChatGPT Work, a powerful agentic desktop productivity tool clearly designed to be OpenAI’s answer to Claude Cowork. This innovative application allows the AI to access your desktop and browser, performing a wide array of tasks for you. From creating spreadsheets and PowerPoints to building entire websites, ChatGPT Work is engineered to streamline complex projects by breaking them down into manageable steps and working on them for hours.
This isn’t an entirely new concept for OpenAI, as agentic capabilities were previously embedded within their Codex app, a tool primarily for programmers. However, in a savvy branding move, OpenAI has separated this agentic work assistant from its coding tool. This repositioning makes ChatGPT Work significantly more appealing and accessible to a broader audience, moving beyond just developers.
One of the most striking demonstrations showcased ChatGPT Work reorganizing an OpenAI employee’s Apple Notes. While this level of automated reorganization might sound daunting to some, it powerfully illustrates the tool’s capabilities. OpenAI encourages users to start with tasks they already know well, allowing them to follow progress, answer questions, change direction, and approve important actions.
Early users are already seeing significant benefits. Angela Ferrante, Head of Enterprise Marketing at Zapier, highlighted how her team leveraged ChatGPT Work to build a repeatable system for reviewing thousands of leads monthly. The AI traced customer touchpoints across Zapier’s CRM, email, and other tools, pinpointed follow-up breakdowns, and generated a weekly executive dashboard that revealed missed pipeline and identified “seven figures in potential sales.”
Much like Claude Cowork, ChatGPT Work boasts extensive connectivity through plugins and connectors. It seamlessly integrates with popular business applications such as Slack, Microsoft Teams, Google Drive, SharePoint, email, calendars, CRMs, and project trackers. This broad integration makes it a versatile addition to almost any professional workflow.
A Smarter, Faster, More Affordable AI Future?
ChatGPT Work is rolling out incrementally, first to Pro, Enterprise, and Edu plans, with Plus and Business subscribers gaining access in the coming days. The potential impact of this tool on daily productivity, especially at the $20/month Plus tier, is something many are eager to test and observe.
OpenAI is also streamlining its desktop application strategy by combining the separate ChatGPT and Codex apps into one unified experience. While it’s not yet clear if ChatGPT Work will be fully folded into this newly combined app, it mirrors Anthropic’s approach of consolidating chat, code, and coworker functionalities. This consolidation points towards a more integrated and user-friendly experience across the OpenAI ecosystem.
The aggressive positioning of GPT-5.6 on benchmarks and cost-effectiveness, coupled with the powerful agentic capabilities of ChatGPT Work, marks a pivotal moment for OpenAI. This comprehensive update aims to redefine efficiency and productivity in the AI space. It begs the question: will GPT-5.6’s faster performance and lower cost make you more likely to choose OpenAI over Anthropic for your AI-powered tasks?
Source: ZDNet – AI
