Key points
- OpenAI released GPT-6.2 Astra and GPT-6.2 Sol to Plus, Pro, Enterprise and Edu users in ChatGPT Work and Codex, and via the API.
- Astra API rates fell 20% to $8 input and $40 output per million tokens; cached input is $0.40. Sol stays at $2 and $10.
- Artificial Analysis ranks Astra second on its Intelligence Index behind Claude Fable 5.2, and says Sol matches September's Astra at about 15% of the cost per task.
- Artificial Analysis found no coding gains and high hallucination rates on hard factual questions; Vals AI found legal weak spots.
Why it matters for marketers: Astra gets cheaper per task and Sol keeps its prices, so teams running agents for slides, documents and research can rerun cost comparisons, though independent tests show mixed results.
OpenAI has released GPT-6.2 Astra, its top model for demanding professional work, and GPT-6.2 Sol, a cheaper model for everyday tasks. Both are live in ChatGPT Work and Codex for paid tiers, and developers can call them in the API. Astra's API prices dropped 20%, with a deeper 60% discount on cached input. Sol keeps its predecessor's rates.
OpenAI says Astra produces slides, spreadsheets and documents that follow supplied templates and professional conventions more closely, down to font sizes. For content teams, that is the practical pitch: fewer edits to decks and reports. OpenAI also reports gains on multistep work benchmarks at lower cost than the previous Astra and Anthropic's Claude Fable 5.2.
Independent results are more mixed. Artificial Analysis places Astra second on its Intelligence Index, behind Fable 5.2, while saying OpenAI leads on cost versus performance. It also reports flat coding scores and high hallucination rates on very hard factual questions. Vals AI found Astra and Sol ranked well below the top on legal research tests. Latent Space noted unusually negative feedback from some influencers, while others blamed their testing setup.
OpenAI also rated Astra High for AI self-improvement and both models Critical for cybersecurity under its own framework, and it added tighter safeguards that restrict some legitimate security testing.
Marketing teams should test Sol first on routine drafting and agent workflows, since it is the cheaper option and OpenAI says it comes close to Astra on some benchmarks. Check Astra against your own templates and fact-checking steps before moving high-stakes work over.
Sources
This brief is our own summary. Read the full reporting at the source.
- OpenAI: Introducing GPT-6.2 Astra and Sol
- The Decoder: OpenAI launches GPT-6.2 Astra and Sol with lower API prices and some benchmark gains
- Latent Space: [AINews] GPT-6.2 Astra, the #2 Frontier Model is better and 20%-600% cheaper, but has mixed reviews; 6.2 Sol likely main workhorse
- TLDR AI: GPT-6.2 Astra 🧠, Microsoft’s seven models 🚀, Replit Missions 🤖
- MarkTechPost: OpenAI Releases GPT-6.2 Astra and Sol With a 20% Cheaper Astra API