OpenAI publishes a model guide for its GPT-6 family

OpenAI has published a new guide outlining how developers should use its GPT-6 model family, including new tools designed to let AI agents work across tasks that can take hours or even days.

The company divides the lineup into three primary models: GPT-6 Astra, GPT-6.1 Sol and GPT-6 Luna, each targeting a different balance of intelligence, cost and speed.    

GPT-6 Astra is positioned as OpenAI’s most intelligent model and is intended for the hardest reasoning workloads. GPT-6.1 Sol targets complex coding, research and computer-use tasks, while Luna is designed for high-volume workloads with clearer objectives such as extraction, classification and structured summaries. 

OpenAI’s current API documentation similarly describes Astra as its highest-intelligence option and Sol as a lower-cost alternative with near-Astra performance.

The models also come with sharply different pricing. Astra costs $10 per million input tokens and $50 per million output tokens. GPT-6.1 Sol costs $2 and $10, respectively, while Luna costs $0.10 for input and $0.50 for output.

OpenAI is also emphasizing longer-running autonomous work with the GPT-6 generation.

The company said the models can now take on tasks spanning hours or days, supported by features including mid-run steering, asynchronous tool calls and parallel work. 

Developers can update instructions while a task is running without canceling existing tool calls, while agents can continue independent work while slower tools operate in the background.

GPT-6.1 Sol also supports multi-agent workflows in beta through the Responses API. The model can delegate independent portions of a task to subagents and combine their findings into a final response.   

GPT-6 Astra extends that approach in Codex, where it can ask users for clarification while continuing portions of a task that do not depend on the answer. OpenAI recommends explicitly defining which decisions an agent can make independently and which require approval.    

All three models can also use computer-control capabilities to interact with websites and desktop applications when direct APIs or connected tools are unavailable. OpenAI said this can allow an agent to investigate a software bug, modify code and then open the product in a browser to verify the fix.

The guide also recommends developers make greater use of prompt caching and context compaction to control the cost of increasingly long agent workflows. Cached input can cost as much as 95% less than uncached tokens depending on the model,

OpenAI also said developers may need to simplify how they prompt newer models. The company argues that increasingly capable systems are better at handling nuance and ambiguity, meaning overly prescriptive instructions can sometimes hurt performance rather than improve it.