← All models

GPT-6 Astra

Published Publisher OpenAIgpt-6-astra

Overview

GPT-6 Astra is OpenAI’s model for demanding end-to-end work. OpenAI lists complex reasoning, coding, computer use, research, and document creation among its intended uses. It can assist with multistep workflows such as online research, software testing, data analysis, and producing documents, spreadsheets, and presentations. (developers.openai.com)

Capabilities

  • Accepts text and image inputs and produces text outputs.
  • Has a 1,050,000-token context window and a maximum output of 128,000 tokens. Its knowledge cutoff is April 30, 2026.
  • Supports reasoning effort levels low, medium, high, xhigh, and max, as well as structured outputs and function calling.
  • Supports tools including web search, file search, code interpreter, computer use, and image generation through the Responses API. (developers.openai.com)

Limitations

  • Audio and video are not supported as model input or output; fine-tuning is not supported.
  • The model does not support none reasoning effort, custom temperature or top_p values, or log probabilities. Tool calling requires the Responses API.
  • Safety checks can pause or stop legitimate work; in the API, a paused task stops. (developers.openai.com)

Release

Released September 3, 2026, according to the OpenAI API changelog. (developers.openai.com)

OpenAI API pricing

The following are OpenAI direct API text-token prices in USD per 1 million tokens, checked November 16, 2026. Standard rates apply to prompts with up to 272K input tokens:

Processing tier Input Output Cached input 30-minute cache write 12-hour cache write
Standard $10.00 $50.00 $1.00 $12.50 $25.00
Flex / Batch $5.00 $25.00 $0.50 $6.25 $12.50
Fast $20.00 $100.00 $2.00 $25.00 $50.00
Ultrafast $60.00 $300.00 $6.00 $75.00 $150.00

For prompts exceeding 272K input tokens, input and cache rates are doubled and output rates are multiplied by 1.5 for the entire request. Eligible regional processing and FedRAMP endpoints carry a 10% uplift. Tool use can incur additional charges; for example, web search is $10 per 1,000 calls plus search-content tokens at model rates. Other platforms may have separate terms. (developers.openai.com)

Official sources