Enterprise DNAEnterprise DNA
LModelsBest vision
OpenAI logo

GPT-6 Astra

by OpenAI

GPT-6 Astra is OpenAI's flagship model for demanding end-to-end work. It is suited for advanced analysis, software engineering, deep research, scientific work, and document creatio

Visit OpenAISubmit your build →
GA

Models

GPT-6 Astra

Added 5 Sept 2026

#openai

Overview

GPT-6 Astra is OpenAI's flagship model for demanding end-to-end work. It is suited for advanced analysis, software engineering, deep research, scientific work, and document creation, with particular strengths in long-horizon...

Best for

Best for
Demanding analysis and engineering

How to use / API access

Call via OpenRouter with model id `openai/gpt-6-astra`.

  • openrouter · openai/gpt-6-astra Docs

OpenRouter id: openai/gpt-6-astra

Benchmarks

  • Intelligence Index (artificial-analysis): 52.7 index
  • Coding Index (artificial-analysis): 76.9 index
  • Agentic Index (artificial-analysis): 51 index

Notes

GPT-6 Astra is listed in the EDNA Models directory from the OpenRouter catalogue.

Pros

  • Analysing competing technical constraints
  • file, image, text input; text output on the listed OpenRouter route

Cons

  • Using a flagship model for every trivial task
  • Presenting uncited generated explanations as verified research
  • Catalogue base token rates have additional pricing overrides. Context length, time window, caching, reasoning, tools or media can change total cost; the simple calculator is not a full quote.

In the AI Pulse

Recent daily briefs from the Enterprise DNA AI Pulse that mention this. The short daily read on what is actually happening in AI.

See today's AI Pulse →

Free 27-page guide

Get the free Developer’s Field Guide

A 27-page field guide to the AI coding workflow with Claude. Claude Code, MCP servers, the prompt patterns that work, and what to delegate. Free.

Enter your work email. We send it straight over, plus a few short notes worth knowing. Unsubscribe any time.

No spam. Unsubscribe any time.

← Back to ModelsSubmit your own entry →

USAGE / OPENROUTER

How much GPT-6 Astra is used

Tokens processed through OpenRouter each day. It shows what developers actually run, not every company's private traffic.

Tokens, last 7 days
370B ▼36.6%
Requests, last 7 days
6.5M
Tool calls, last 7 days
3.7M
Intelligence score
52.7
People's vote
1443
Average cost per request
$24.32
Price per million tokens
$10 in · $50 out
0250B500B19 Sept: 140B tokens, 1.7M requests20 Sept: 109B tokens, 2.1M requests21 Sept: 162B tokens, 9.5M requests22 Sept: 145B tokens, 2.1M requests23 Sept: 173B tokens, 2.1M requests24 Sept: 140B tokens, 1.6M requests25 Sept: 143B tokens, 2.3M requests26 Sept: 182B tokens, 2.2M requests27 Sept: 88.4B tokens, 1.8M requests28 Sept: 113B tokens, 2.3M requests29 Sept: 83.2B tokens, 1.8M requests30 Sept: 157B tokens, 2.2M requests1 Oct: 460B tokens, 5.5M requests2 Oct: 429B tokens, 4.6M requests3 Oct: 83.8B tokens, 1.4M requests4 Oct: 83.3B tokens, 1.9M requests5 Oct: 92.3B tokens, 1.5M requests6 Oct: 74B tokens, 1.4M requests7 Oct: 86.8B tokens, 1.8M requests8 Oct: 84.4B tokens, 1.6M requests9 Oct: 98.7B tokens, 1.7M requests10 Oct: 82.1B tokens, 1.2M requests11 Oct: 64.7B tokens, 886K requests12 Oct: 74.2B tokens, 1.1M requests13 Oct: 61.4B tokens, 1.1M requests14 Oct: 45.8B tokens, 931K requests15 Oct: 49.9B tokens, 860K requests16 Oct: 61.8B tokens, 1.2M requests17 Oct: 38.7B tokens, 689K requests18 Oct: 38B tokens, 633K requests19 Sept3 Oct17 Oct18 OctSee where it ranks on the leaderboard →

MODEL FIELD GUIDE / SOURCE EVIDENCE

What you can do with GPT-6 Astra.

Specifications reviewed; no EDNA task evaluation has been run. Sources checked 2026-10-20. Hosting routes and app features can differ.

DeveloperOpenAI
Exact model identifieropenai/gpt-6-astra
Inputfile, image, text
Outputtext
Tool calling
Listed support
Structured output
Listed support
Reasoning controls
Listed support
Availability
available
Price basis
token
Licence verification
Check the developer licence; the model name is insufficient

Catalogue base token rates have additional pricing overrides. Context length, time window, caching, reasoning, tools or media can change total cost; the simple calculator is not a full quote.

Consider it for

  • Analysing competing technical constraints
  • Preparing a research plan from supplied source material

Watch for

  • Using a flagship model for every trivial task
  • Presenting uncited generated explanations as verified research

PRACTICAL EXERCISE / NOT RUN BY ENTERPRISE DNA

Try the model on a small, checkable task.

A queue handles 120 requests per minute but receives 150 for 20 minutes. Calculate backlog, then calculate clearance time at 90 incoming per minute. State assumptions.

Check the result: Backlog is 600 requests and clearance takes 20 minutes under constant service and arrival rates.

Compare another approach

anthropic claude sonnet 5 5 ↗openai gpt 6 1 sol ↗
Inspect the sources and their scope
provider source ↗

Developer documentation for model family or task. OpenRouter route-specific properties remain catalogue claims; EDNA has not executed a task evaluation.

Checked 2026-10-19
catalogue source ↗

openai/gpt-6-astra: listed route capabilities, availability and prices; provider endpoints can differ.

Checked 2026-10-20
How to evaluate a model for your work →

FROM SPECIFICATIONS TO YOUR WORK

Is this the right model for your job?

Start with a realistic sample. These workflows match the model’s listed capabilities, not a verified result on your task.

Write a clear business documentWorkflow & starter prompt ↗Analyse a spreadsheetWorkflow & starter prompt ↗Research a companyWorkflow & starter prompt ↗Build an appWorkflow & starter prompt ↗Extract document dataWorkflow & starter prompt ↗Run a reliable agentWorkflow & starter prompt ↗

Check the source

Price date: 2026-10-20. Prices are USD API token rates. Your app subscription, uploads, browsing and tools are separate.

Model source & availability ↗

Understand the evidence

Listed capabilities and third-party benchmark scores do not establish Enterprise DNA task-test results. Validate accuracy and total cost on your own sample.

Intelligence Index: 52.7 (2026-10-18) ↗Coding Index: 76.9 (2026-10-18) ↗Agentic Index: 51 (2026-10-18) ↗
Compare this model with alternatives →