OpenAI Begins Limited Preview of GPT-5.6 Series with Sol, Terra, and Luna Models
OpenAI has begun a limited preview of its GPT-5.6 model family, which includes the Sol, Terra, and Luna variants, announced on its website. The preview is available to a small group of trusted partners before a broader rollout in the coming weeks.
GPT-5.6 Sol is described as the company’s most capable model, designed for advanced reasoning and cybersecurity tasks. It introduces a new max reasoning effort feature to allow deeper analysis and an ultra mode that coordinates subagents for complex tasks. Terra is positioned as a balanced model for general work, offering similar performance to GPT-5.5 at half the cost, while Luna provides faster responses at the lowest price tier.
All three variants include strengthened safeguards against misuse, with layered protections that combine model-level training, real time monitoring, and account-based review. OpenAI said it dedicated over 700,000 GPU hours to automated red teaming to identify universal jailbreaks and improve robustness against adversarial attacks.
Pricing for the models is set per one million tokens: Sol at $5 for input and $30 for output, Terra at $2.50 and $15, and Luna at $1 and $6. During the preview, access is available through the API and Codex, with plans to expand to ChatGPT users soon. OpenAI also plans to launch GPT-5.6 Sol on Cerebras hardware in July, allowing processing speeds of up to 750 tokens per second.
We hope you enjoyed this article
Consider subscribing to one of our newsletters like Cybersecurity AI Weekly or Daily AI Brief.
Also, consider following us on social media:
More from Cybersecurity
Oct 16 Inlayer, JLINC and Vconic Introduce Human Verification for Calls Oct 16 CTM and Huawei Plan AI Network Security Upgrades in Macao Oct 16 Advantech and DEKRA Partner on Industrial Cybersecurity Compliance Oct 15 Cyfendo Offers AI Code Scanning and Patch Review for Small Businesses Oct 15 Perplexity Open Sources SPACE Agent SandboxCybersecurity AI Weekly
Weekly newsletter about AI in Cybersecurity.
Whitepaper
Tensordyne Napier: What If One Rack Could Do the Work of Nine?
Tensordyne
This Tensordyne whitepaper presents Napier, an inference-focused AI processor and rack-scale system based on the company’s TDN Math logarithmic number system. It examines infrastructure requirements for large mixture-of-experts and agentic models, compares major inference architecture approaches, and details the TDN AIP processor, TDN72 pod, TDN Link fabric, and Napier Ultra configuration. The paper reports simulation-based performance, cost, and accuracy-validation results, including Tensordyne’s projected comparison of one Napier rack with a nine-rack Nvidia Rubin plus Groq deployment; the chip is reported as taped out and in fabrication.
Read moreYou may also like
OpenAI Releases GPT-6.1 Sol at One Fifth of Astra Token Pricing
OpenAI Rolls Out GPT-6 and Intelligent UI in ChatGPT
OpenAI Previews Decisions API for Faster Model Choices
OpenAI to Retire GPT-4o and GPT-4.1 API Models in January
OpenAI Offers US University Labs Free GPT-Rosalind Access
Daily AI Brief: the AI news that matters, in your inbox.