OpenAI Introduces GPT-6.2 Astra and GPT-6.2 Sol Models
OpenAI has introduced two new models, GPT-6.2 Astra as a frontier model for complex projects and GPT-6.2 Sol as an efficient option for daily tasks. Developer Simon Willison immediately integrated support for these models into version 0.38 of his LLM utility tool. These new models offer improved pricing structures, with GPT-6.2 Astra priced at 80% of its predecessor and offering significantly cheaper context caching. This reduction makes long-running agentic AI sessions and complex developer workflows much more cost-effective. GPT-6.2 Astra costs $8 per million input tokens and $40 per million output tokens with a cached price of $0.40 per million, matching competitors like Claude Fable 5.2. Meanwhile, GPT-6.2 Sol retains the flat pricing of $2/$10 per million tokens.
## BACKGROUND
Simon Willison's LLM is a popular open-source command-line utility and Python library that allows developers to interact with various large language models through remote APIs or local environments. Agentic AI refers to autonomous systems capable of planning, utilizing tools, and executing complex workflows over extended sessions without constant human intervention.