It's been a busy few days (very busy, in fact) at OpenAI, which seems to have finally responded to the "code red" activated by its CEO, Sam Altman, at the end of last year. In the same week that we saw the launch of Images 2.0 and Workspace Agents, a new version of GPT, 5.5, has arrived.

5 key points about the new GPT 5.5

1. Solving complex problems

Specifically, the company seems very proud of this new model's ability to solve complex problems composed of multiple related subtasks. The main idea of ​​the launch announcement is that GPT 5.5 is able to plan, use different tools, verify its work, and continue until it finds the solution to the core of the assigned problem, without having to specify the subtasks to be solved.

And all this "often achieving higher quality results with fewer tokens and fewer retries."

2. GPT 5.5 and Codex

According to published metrics, the new model stands out for its significant leap in coding and computer control. OpenAI has presented metrics that place it at the forefront, achieving a 58.6% success rate in resolving real GitHub issues and reaching 82.7% accuracy in console tasks and tool coordination. Furthermore, the model is capable of operating software directly in real-world computing environments, allowing it to manage office workflows such as creating documents, spreadsheets, and presentations, acting more like an executive assistant than a simple chatbot.

3. A Model Geared Towards the Agent Era

It makes perfect sense that the presentation of GPT 5.5 coincided with the release of Workspace Agents, as this new model appears to be clearly designed to fuel the boom in AI agents. As we mentioned earlier, integrating the model with collaborative tools allows AI to not only draft, summarize, or analyze information, but also to handle longer tasks: searching for data, using tools, verifying results, and transforming disorganized materials into documents, spreadsheets, presentations, or work plans.

4. More Efficient and Secure

GPT-5.5 is smarter and more efficient. OpenAI explains that it has improved token generation speed by 20% and is more efficient, especially in programming tasks, where it can reduce token usage, lowering operating costs compared to competing models. This is complemented by the introduction of GPT-5.5 Pro, a superior version for demanding environments in the legal, scientific, and data analytics sectors, where accuracy and complex reasoning are critical.

Finally, OpenAI emphasizes that this model comes with its most stringent security systems to date, reinforcing it in sensitive areas such as cybersecurity and biology. The idea is for AI to be a massive, yet secure, productivity tool—a key consideration amidst the controversy surrounding Anthropic's Mythos model, which has been deemed a global security risk.

5. Availability and Pricing

The standard GPT-5.5 model is already available to ChatGPT Plus, Pro, Business, and Enterprise users. The GPT-5.5 Pro version, focused on maximum accuracy and complex tasks, is reserved for the Pro, Business, and Enterprise tiers. Regarding its integration with Codex, it will offer a context window of 400,000 tokens for educational and enterprise plans, including a fast mode that increases generation speed at a higher cost.

As for the API, developers will be able to access GPT-5.5 with a context window of one million tokens. The base prices will be $5 (entry) and $30 (exit) per million tokens, while the Pro version will raise these fees to $30 and $180 respectively. Although the unit price is higher than in previous versions, the model consumes fewer tokens to perform the same tasks, thus balancing the final cost for the user thanks to its greater intelligence and optimization.