OpenAI announced on Wednesday that the latest iteration of its flagship chatbot, the ChatGPT-5.6 family, is now live. The rollout includes three distinct models—Sol, Terra and Luna—each tuned for a different set of user needs. OpenAI says the models are being deployed globally and should be accessible to all users within the next 24 hours.
Sol, positioned as the premium offering, is built for complex reasoning, end‑to‑end technical execution and high‑level design work. The company claims Sol is its strongest coding model to date, outpacing Anthropic’s Fable 5 in the Artificial Analysis Coding Agent Index. Benchmarks also show Sol delivers better efficiency across token output, processing time and cost.
Terra occupies the middle of the lineup, aimed at everyday workflows. OpenAI markets it as the go‑to choice for most users who need solid performance without the premium price tag of Sol. Luna, the third variant, prioritizes cost efficiency and speed, making it attractive for high‑volume, low‑latency applications.
Beyond raw performance, the 5.6 series brings new design capabilities. The models can now inspect rendered visual outputs and flag issues, a step forward for creators who rely on AI‑generated graphics. OpenAI also highlighted the series’ cybersecurity strengths, noting that the models achieve high performance with fewer tokens, a factor that regulators have scrutinized.
Because of those security considerations, OpenAI initially limited the release to cybersecurity and technology professionals, a move echoed by rivals such as Anthropic. In Anthropic’s case, the U.S. government intervened, barring foreign nationals from accessing its latest model while officials evaluated safety measures. OpenAI’s staggered rollout follows a similar precautionary approach, aligning with recent guidance from government officials.
ChatGPT Work expands AI‑driven automation
Alongside the model family, OpenAI unveiled ChatGPT Work, an AI agent that lives inside the ChatGPT interface. Powered by the new 5.6 models, the agent can pull information from a user’s apps and files, devise a plan and execute it with little or no further interaction. Users can also set up scheduled tasks, allowing the agent to operate while they are away from their computers. The feature is designed to streamline repetitive workflows and give users a hands‑free way to manage digital tasks.
The combined launch underscores OpenAI’s strategy of pairing ever‑more capable models with practical tools that bring those capabilities into everyday productivity. By offering a tiered model suite and an automation agent in a single release, the company aims to serve developers seeking cutting‑edge performance, businesses looking for cost‑effective AI, and individuals who want AI assistance without deep technical expertise.
OpenAI’s announcement comes nearly four years after the original ChatGPT debuted, a period during which the firm has maintained a lead in user numbers despite growing competition from Google, Anthropic and other AI firms. As the 5.6 models roll out, the company will likely monitor adoption patterns and regulatory feedback closely, especially given the heightened focus on AI safety and security.
This article was written with the assistance of AI.
News Factory APP - agentic news to boost your SEO & AEO.