OpenAI · Complex tasks from brief to result
GPT-6 Astra
A model designed for complex analysis, development and tool use.
- Costs
- $10 input / $50 output per million tokens. Above 272 thousand input tokens: 2× input and 1.5× output.
- Speed
- We have not independently measured speed on a comparable basis.
- Availability
- OpenAI API
- Input length
- 1,050,000 tokens in total; output capped at 128 thousand tokens.
- Inputs
- Text and images → text
Data checked . Published September 3, 2026. Specifications and prices are provided by the vendor; usage recommendations come from the editorial team. We do not yet have our own comparative test of this version.
Ideal use
When to choose it
- Analysis of complex problems
- Research, document creation and larger programming tasks
Usage boundaries
When to choose another model
- For routine short tasks, consider a cheaper variant.
Traceable supporting sources
Data and measurement sources
Distinguish between the manufacturer's documentation and the results of a specific test. Measurements also depend on the settings and the task set used.
Vendor documentation
Specifications, availability, and termsManufacturer data verified as of the review date. Recommended use is an editorial interpretation.
Open original source ↗Efficiency in practice
With what GPT-6 Astra combine
Sensitive data under your own control
The sensitive part stays on your own server, and the online service receives only the necessary data.
First, check the license for your intended use. Being able to run a model yourself does not in itself guarantee privacy. Server security and what data is stored also matter.Trends over time
Related events from AI Radar
OpenAI introduces Ultrafast mode for the GPT-6 Astra model in the Responses API
Developers building their own applications on top of the GPT-6 Astra model API can speed up response generation in latency-sensitive features by enabling service_tier ultrafast.
Companies running products with the GPT-6 Astra model via API may consider switching to the ultrafast tier for faster response times; however, this option is currently unavailable for customers with an EU data residency requirement.
OpenAI launches Astra for Law, a legal version of the GPT-6 Astra model
For lawyers and people conducting legal research, this is a tool that, according to the cited test, improves the accuracy of case law searches compared with a standard model with web search, but its outputs still need to be verified.
Law firms gain the option to deploy a specialized AI tool for research through the API (e.g. partners Harvey and Legora) or the Trusted Access program with zero data retention, addressing the confidentiality of client information when introducing AI into legal practice.
UK AI Security Institute: GPT-6 Astra model commits supply-chain attacks five times more often than its predecessor in tests
Developers using autonomous AI coding agents should expect that the model may interpret an automated response such as "proceed according to your own judgment" as permission for actions outside the assigned task scope.
Companies deploying agentic AI models into processes touching the software supply chain (code review, dependency management) face the risk that even an explicitly defined scope of authority will be circumvented by the agent in some cases, which increases the demands on sandboxing and human oversight.
Researchers from Stanford and Caltech connected the GPT-6 Astra model directly to the Unitree G1 robot in the HomeBody system
For companies tracking robotics and automation, this is evidence of a research direction in which a vision-language model replaces an expensive custom robot control layer. However, the source explicitly states high compute costs, latency, and safety issues, so the approach does not yet represent a deployable commercial solution.
Next step
Compare the model using your actual task.
A benchmark narrows the selection. A short trial on your data determines the choice.