OpenAI is preparing to release GPT-5.1, as indicated by its presence on the ChatGPT website. This new model, particularly labeled "Thinking," is expected to enhance advanced reasoning capabilities within ChatGPT, potentially improving its performance on multi-step tasks. Along with this model, there are rumors of a broader lineup that includes variations like Mini and a Codex-focused upgrade, each tailored for different user needs and computational limits. These versions are believed to be in internal testing, often under obscure names, suggesting A/B testing is underway before a public rollout. The timing of these developments is significant. OpenAI seems to be aligning its release schedule with Google’s upcoming Gemini 3 Pro, likely to maintain competitiveness in the AI landscape. Both companies are working on models that offer larger context windows and better problem-solving skills. Reports mention that these new models might feature varying “thinking budgets,” which could affect their context length and reasoning depth. This will cater to a range of users from casual consumers to enterprise teams looking for dependable AI outputs. While the leak hints at exciting advancements, it's uncertain whether all the rumored models will be released at the same time or if some will be limited to specific tiers or API access initially. The details emerged from a close examination of the web application’s backend, a common method for spotting upcoming features. TestingCatalog will continue to keep an eye on any official updates from OpenAI regarding the capabilities and release dates of these models, especially with the competitive pressure rising ahead of Gemini 3 Pro's launch.