AI KNOWLEDGE DESK

Models · entities · concepts · comparisons · practical tools

GETLLMS.ORG
ModelFrontier agent models

GPT-6 Astra

GPT-6 Astra is OpenAI's September 3, 2026 flagship model with the API ID gpt-6-astra, a 1,050,000-token context window, 128,000-token maximum output, text and image input, and $10/$50 per-million-token base pricing. It has been released, but it is not yet generally available: Trusted Access organizations receive it first and broader API and plan access is rolling out over the following days.

Why it matters

Astra is the first OpenAI model positioned for the hardest complete workflows rather than a single reasoning step. It combines high reasoning effort, computer use, hosted tools, document creation, and long context, but its staged access, high token price, long-context surcharge, and Critical cybersecurity classification make availability and governance part of the model-selection decision.

Source-backed summary

OpenAI's API page controls the model ID, limits, modalities, pricing, tools, and rollout wording. OpenAI's release notes and safety overview establish that access is still limited and that Astra is the company's first model to reach the Critical cybersecurity capability threshold. Artificial Analysis provides a dated independent benchmark view; it does not override OpenAI's product contract.

Primary use cases
  • Run difficult end-to-end coding and repository tasks with tools and verification.
  • Coordinate research, computer use, and document creation in one long workflow.
  • Escalate tasks that fail on cheaper models after measuring the value of the higher price.
  • Evaluate high-capability cyber-defensive work only inside approved access and governance boundaries.
Released does not mean generally available

OpenAI released GPT-6 Astra on September 3, but the current product language says Trusted Access enterprises receive it first and broader API, Plus, Pro, Business, and Enterprise access will arrive over the coming days. Teams should check the model picker or API account instead of assuming every paid account can call it now.

API, context, and price

The documented API model ID is gpt-6-astra. It accepts text and image input, returns text, supports low through max reasoning effort, and provides a 1,050,000-token context window with up to 128,000 output tokens. Base rates are $10 input, $1 cached input, $12.50 cache write, and $50 output per million tokens.

  • Requests above 272,000 input tokens use 2x input and cache rates plus 1.5x output rates for the full request.
  • Batch and Flex cost 50% of standard rates; Fast mode costs 2x the applicable rate.
  • Responses API tools include web and file search, image generation, code interpreter, hosted shell, apply patch, skills, computer use, MCP, and tool search.
Capability, cost, and safety are separate questions

OpenAI calls Astra its most capable model, while independent September 3 testing shows strong coding-agent gains but a mixed value story on a broader intelligence index because the model costs 2.5 times GPT-5.6 Sol at current promotional rates. OpenAI also classifies Astra at the Critical cyber threshold, so organizations should validate authorization boundaries, monitoring, and review workflows alongside task quality.

GPT-6 Astra FAQ

Common questions about GPT-6 Astra.

Is GPT-6 Astra generally available?+

No. OpenAI released Astra on September 3, 2026, but current access is rolling out first to Trusted Access organizations. Broader API and paid-plan availability is planned over the following days, so check the active account rather than assuming access.

How much does GPT-6 Astra cost?+

Standard API pricing is $10 per million input tokens, $1 cached input, $12.50 cache writes, and $50 output. Requests above 272,000 input tokens cost more for the whole request, while Batch and Flex are half price and Fast mode is double price.

Is GPT-6 Astra the best AI model?+

Not for every workload. Astra is a strong choice for difficult OpenAI-centered agent and coding workflows, but Claude Fable 5.1 leads the current independent composite at max effort, Gemini 3.8 Flash is far cheaper and faster, and access to Astra is still limited. Use the same task harness and acceptance checks before choosing.