OpenAI has updated GPT-5.6 Sol with improvements to factual accuracy and response focus. In internal evaluations on financial, medical, and legal prompts requiring factual detail, responses containing at least one factual error dropped 68% compared to GPT-5.5 Instant, and 62% lower than prior baselines. The model now adapts response length to question complexity—shorter direct answers for quick queries, fuller context for research and planning. A new think-slider for Plus/Pro users lets ChatGPT allocate computation depth per query, with a separate 'Think' button for free users to access higher reasoning on harder problems.
The improvements target ChatGPT's primary use case: roughly 1 billion weekly users across quick Q&A, web search, research, planning, and decision-making. GPT-5.6 Luna (lower compute) became the free-tier default with unlimited text chats, while Plus/Pro users get Sol as both instant and deeper-reasoning backbone. OpenAI emphasizes consistency across reasoning modes—prior Instant and Thinking responses now use the same foundation model, reducing context-switching overhead.
For practitioners deploying ChatGPT in production pipelines (customer support, technical Q&A, grounded retrieval workflows), the 68% factual-error reduction and tighter formatting discipline matter. OpenAI did not publish absolute error rates on real-world tasks or external benchmarks, so gains should be validated in your use case before relying on Sol for regulated domains (finance, healthcare, legal). The free-tier expansion and expanded access suggest OpenAI is optimizing for scale and default reliability rather than frontier capability.