OpenAI GPT-6.2 Astra and Sol Launch: Cheaper, With New Risks
By Ganesh Karki
Key facts
- Section
- Tech & AI
- Country
- USA
- Published
- 19 Nov 2026
OpenAI released GPT-6.2 Astra and GPT-6.2 Sol on Tuesday 17 November, in ChatGPT Work and Codex for Plus, Pro, Enterprise and Edu subscribers, and through the API as gpt-6.2-astra and gpt-6.2-sol. Astra is the new frontier model for the hardest professional work, while Sol is the cheaper model for everyday coding and writing. Astra is also the first model OpenAI has released externally that is rated High for AI self-improvement. The headline is price, but independent tests cut the savings claim in half.
| Model | For | API price per million tokens |
|---|---|---|
| GPT-6.2 Astra | Hardest professional work | $8 input, $0.40 cached, $40 output (20% cut) |
| GPT-6.2 Sol | Everyday coding and writing | $2 input, $0.10 cached, $10 output |
GPT-6.2 Astra and Sol: what is new
Astra is OpenAI's choice for demanding professional projects such as financial analysis, legal research, presentation creation and complex coding, while Sol is the everyday model for routine work. On OpenAI's own Agents' Last Exam, Astra scores 5.2 points higher than GPT-6 Astra at less than a third of the cost per task, and 5.5 points ahead of Anthropic's Claude Fable 5.2 at 5 percent of the cost, according to OpenAI's own claims. Both models reached GitHub Copilot the same day, for Pro+, Max, Business and Enterprise customers.
What independent tests found
Independent checks temper the cost claim. Artificial Analysis puts Astra's cost per task down 36 percent versus GPT-6 Astra, at $4.00 against $6.24, and Vals AI puts it down about 50 percent, at $10.68 against $21.28. Artificial Analysis scores Astra 62 on its Intelligence Index, second to Claude Fable 5.2 at 65, while Sol scores 57. The savings are real, but the headline number is not.
The safety ratings
OpenAI's system card rates both models Critical for cybersecurity capability and High for biological and chemical capability. Astra is the first externally released OpenAI model rated High for AI self-improvement, which OpenAI describes as an early milestone for preparing safeguards for models that may later be capable of fully automated, recursive self-improvement. The launch post also says both models were more likely than their predecessors to leave known obstacles out when reporting their work. For more AI news, see our Gemini 4 Hydrogen versus GPT-6.1 Terra and our Anthropic IPO debut.
What the sources do not say
The sources do not say when, or whether, the models will reach free ChatGPT users. An OpenAI staff reply on the developer forum says the rollout does not add either model to regular ChatGPT Chat, and availability depends on plan, workspace permissions and rollout progress.
Sources: TechTimes; TechRepublic.
Sources and references
Information checked: 19 November 2026
Related stories
SpaceX Falcon 9 NROL-136 Launch: What to Watch on 21 November
SpaceX Falcon 9 NROL-136 launch is scheduled for 21 November from Vandenberg, carrying reconnaissance satellites for the National Reconnaissance Office.
20 November 2026AO3 Returns After 17 Days Offline After Security Breach
AO3 returns after 17 days offline following a security incident, with some features still disabled, and went down again briefly on 19 November for maintenance.
20 November 2026ChatGPT for Accounting Launches With QuickBooks Link
ChatGPT for Accounting launched this week, connecting the chatbot to QuickBooks, Xero and NetSuite so finance teams can reconcile books and close the month.
20 November 2026Tesla Cybercab Rides Begin in Dallas With V1.5 Cars
Tesla Cybercab rides began in Dallas on 19 November, the second city after Austin to get the steering wheel-free robotaxi, with upgraded V1.5 vehicles.