Model icon
gpt-6.2-astra
Model:
GPT-6.2 Astra is OpenAI's frontier model for the hardest professional projects: slides, spreadsheets, and documents that follow your templates and your field's conventions, scientific analysis carried out step by step in a terminal, and merge-ready changes in large codebases. OpenAI reports it scores 5.2 points above GPT-6 Astra on Agents' Last Exam at less than a third of the cost. Send text and images with a 1.05M-token context and up to 128K output tokens.
Chat

This Playground is for demo purposes. Data is only valid in the current window and will be cleared on refresh.

Type a message...

Configuration
1.0
1.0
Pricing details

Pay as you go. No subscription, no hidden fees.

1 credit = $0.005
GPT-6.2 Astra
Input
$6.40/1M tokens
1280 credits/1M tokens
$8.00/1M tokensSave 20%
Cache read
$0.320/1M tokens
64 credits/1M tokens
$0.400/1M tokensSave 20%
Cache write
$8.00/1M tokens
1600 credits/1M tokens
$10.00/1M tokensSave 20%
Output
$32.00/1M tokens
6400 credits/1M tokens
$40.00/1M tokensSave 20%
How it's billed: Input and output tokens are billed separately, per 1M tokens.

Charges are based on the final output.

GPT-6.2 ASTRA

Frontier reasoning for professional deliverables, science, and complex coding

GPT-6.2 Astra API: OpenAI's Frontier Model with 1M Context

GPT-6.2 Astra is OpenAI's frontier model for the hardest projects and the successor to GPT-6 Astra. It drafts slides, spreadsheets, and documents that follow your templates and the conventions of your field, carries out scientific work step by step in a terminal, and implements and tests complex changes in existing codebases.

The model accepts text and image input, returns text, and supports a 1,050,000-token context window with up to 128,000 output tokens. On PoYo, call it with model ID gpt-6.2-astra through Chat Completions or Responses, and choose a reasoning effort from low to max.

Key Features of GPT-6.2 Astra API

OpenAI built GPT-6.2 Astra for complex professional work, from deliverables that follow your field's conventions to scientific analysis and changes in real codebases, using fewer output tokens per task.

01

Multistep Professional Work at a Lower Cost per Task

Long-horizon professional workflows run through many steps, and every step adds cost. GPT-6.2 Astra matches its tools and effort to the workflow and finishes this kind of work more efficiently. On Agents' Last Exam, which tests multistep professional work, OpenAI reports that GPT-6.2 Astra sets a new high, scoring 5.2 points above GPT-6 Astra at less than a third of the cost.

  • Carry long professional workflows through many steps
  • Match tools and effort to each task
  • Score higher than GPT-6 Astra at lower cost
Agents' Last Exam v1.1 score versus cost per task for GPT-6.2 Astra, GPT-6.2 Sol and other models

02

Professional Documents That Follow Your Templates

AI-drafted slides, spreadsheets, and documents often need heavy editing before they match a team's format. GPT-6.2 Astra follows the templates you provide and the conventions of fields like law, sales, finance, and accounting, with more direct writing and less filler. On OpenAI's internal GDPval v2, built from long-horizon assignments in finance, data science, biotechnology, and marketing, it leads the models tested, up 6.2 percentage points from GPT-6 Astra while using 22% fewer output tokens.

  • Follow provided templates and field conventions
  • Write direct, specific copy with less filler
  • Use 22% fewer output tokens than GPT-6 Astra
GDPval v2 score versus output tokens for GPT-6.2 Astra, GPT-6.2 Sol and other models

03

Scientific Research Carried Out in a Terminal

Scientific work means writing code, running analyses, and interpreting results over many terminal steps. OpenAI believes GPT-6.2 Astra is the best model in the world for scientists. On Terminal-Bench-Science 0.1, which tests how well models carry out scientific work in a terminal, OpenAI reports a new high, 9 percentage points above GPT-6 Astra at less than one-third the cost. It also sets a new high on the Vals Science Index.

  • Run scientific analyses step by step in a terminal
  • Interpret real biological data and experiment results
  • Reach higher scores than GPT-6 Astra at lower cost
Terminal-Bench-Science 0.1 score versus cost for GPT-6.2 Astra, GPT-6.2 Sol and other models

04

Merge-Ready Changes in Real Codebases

A code change is only useful once it meets the maintainers' bar for merging. GPT-6.2 Astra is better at implementing and testing complex changes in existing codebases, and early testers reported cleaner code style, more focused tests, and fewer rounds of revision. FrontierCode 2.0 evaluates whether changes to real codebases meet maintainers' standards for merging, and OpenAI's chart shows GPT-6.2 Astra scoring above GPT-6 Astra at a lower cost.

  • Implement and test changes in existing codebases
  • Write cleaner code with more focused tests
  • Build frontends for websites, games, and 3D scenes
FrontierCode 2.0 score versus cost for GPT-6.2 Sol, GPT-6.2 Astra and other models

Who Can Benefit from GPT-6.2 Astra API?

01

Finance and Consulting Teams

Pull the relevant figures from source files, build valuation workbooks and investment memos, and turn findings into client decks that follow your firm's templates with GPT-6.2 Astra.

02

Legal Teams

Find the relevant facts across source documents, apply them to a detailed brief, and draft memos and contract reviews that follow the conventions of legal work.

03

Scientists and Research Labs

Run scientific analyses in a terminal, interpret experimental and biological data, and write the code that carries a research workflow from raw data to results.

04

Software Engineering Teams

Assign complex changes in existing codebases to GPT-6.2 Astra, with tests included, and review cleaner code that needs fewer rounds of revision.

05

Web and 3D Developers

Turn a written brief into a website, browser game, or interactive 3D scene, with layouts and visuals built around the goal of each project.

06

AI Agent Builders

Build multistep agents with function tools through the Responses API, and set the reasoning effort for each step from low to max.

GPT-6.2 Astra vs GPT-6 Astra — What's Different?

GPT-6.2 Astra keeps GPT-6 Astra's context window, output limit, and input types. OpenAI reports higher scores on multistep professional work, scientific research, and long-horizon knowledge work, with lower cost per task and lower official token prices.

CapabilityGPT-6.2 AstraGPT-6 Astra
Best forThe hardest projects: professional deliverables, science, and complex codingThe hardest end-to-end work
Input / outputText and images in, text outText and images in, text out
Context window1,050,000 tokens1,050,000 tokens
Max output128,000 tokens128,000 tokens
Reasoning effortlow, medium, high, xhigh, maxlow, medium, high, xhigh, max
Agents' Last Exam v1.15.2 points higher at less than a third of the costBaseline
Terminal-Bench-Science 0.19 percentage points higher at less than one-third the costBaseline
GDPval v2 (internal)6.2 percentage points higher with 22% fewer output tokensBaseline
Official price · input / cached input / output per 1M tokens$8 / $0.40 / $40$10 / $1.00 / $50

Benchmark comparisons and official prices come from OpenAI's GPT-6.2 Astra launch announcement. PoYo rates for GPT-6.2 Astra are listed in the pricing table above.

How to Use GPT-6.2 Astra API on PoYo

Step 1: Register and Create an API Key
Sign in to PoYo and create a key for your application.
Create API Key →

Step 2: Add Credits
Fund your account. GPT-6.2 Astra is billed per token at 1280 credits per million input tokens and 6400 credits per million output tokens.
Add Credits →

Step 3: Integrate the API
Send requests with model gpt-6.2-astra to /v1/chat/completions or /v1/responses, with streaming on either endpoint. Use /v1/responses for function tools. Test prompts, image inputs, and reasoning effort in the Playground before moving them into your application.
View Documentation →

Frequently Asked Questions about GPT-6.2 Astra API

What is GPT-6.2 Astra?

GPT-6.2 Astra is OpenAI's frontier model for the hardest projects, released on November 17, 2026, alongside GPT-6.2 Sol. It is built for professional documents, scientific research, complex coding, and agent workflows, accepts text and image input, and returns text. On PoYo, call it with model ID gpt-6.2-astra through Chat Completions or Responses.

How is GPT-6.2 Astra different from GPT-6 Astra?

OpenAI reports that GPT-6.2 Astra scores 5.2 points higher on Agents' Last Exam at less than a third of the cost, 9 percentage points higher on Terminal-Bench-Science 0.1 at less than one-third the cost, and 6.2 percentage points higher on GDPval v2 with 22% fewer output tokens. Its official input and output prices are 20% lower, and cached input is 60% lower.

Should I use GPT-6.2 Astra or GPT-6.2 Sol?

Choose GPT-6.2 Astra for the most demanding projects, such as professional deliverables, scientific research, and complex codebase changes. OpenAI positions GPT-6.2 Sol as its efficient, versatile model for everyday work; on FrontierCode 2.0, Sol comes within 0.5 points of Astra at about one-fifth the cost (both at max effort). Both models are available on PoYo.

How much does GPT-6.2 Astra API cost on PoYo?

PoYo charges 1280 credits ($6.40) per million input tokens, 6400 credits ($32) per million output tokens, 64 credits ($0.32) per million cache-read tokens, and 1600 credits ($8) per million cache-write tokens. OpenAI's standard rates are $8, $40, $0.40, and $10, so PoYo's rates are 20% lower. You pay only for the tokens you use.

What context window and reasoning settings does GPT-6.2 Astra support?

GPT-6.2 Astra supports a 1,050,000-token context window, up to 128,000 output tokens, and a May 18, 2026 knowledge cutoff. Set reasoning_effort to low, medium, high, xhigh, or max; medium is the default, and OpenAI recommends it as the starting point for new integrations.

Can I use GPT-6.2 Astra with Chat Completions and Responses?

Yes. PoYo serves GPT-6.2 Astra on /v1/chat/completions and /v1/responses, and both endpoints support streaming. For function tools, send requests to the Responses API, which OpenAI's GPT-6.2 guide specifies for tool calling. Existing OpenAI SDK clients connect by changing the base URL and API key.

Can GPT-6.2 Astra analyze images?

Yes. GPT-6.2 Astra accepts image input together with text, so you can send screenshots, charts, slide templates, and scanned documents in the same request and get a text response that draws on both.

Why Use PoYo for GPT-6.2 Astra API Access

01

Familiar API Formats

Call GPT-6.2 Astra through OpenAI-compatible Chat Completions or Responses, so existing OpenAI clients can connect with a base URL and key change.

02

One Key for Many Models

Use the same PoYo API key for GPT-6.2 Astra and other chat, image, video, and audio models available on PoYo.

03

20% Below Official Token Rates

Pay per million input, output, and cached tokens at rates 20% below OpenAI's standard prices, with no subscription required.

04

Playground Before Integration

Test prompts, image inputs, and reasoning effort with GPT-6.2 Astra in the Playground, then reuse the same request format in your application.