Back to Models

GPT-6 Astra

openaiopenai/gpt-6-astra

OpenAI's most capable GPT-6 model, built for the hardest end-to-end work.

GPT-6 Astra is OpenAI's newest frontier model, verified live 2026-09-16/17 directly against OpenAI's own API pricing and model-detail pages. TheRouter routes it exclusively through the OpenAI Responses API (matching OpenAI's own guidance that reliable tool calling for this model requires Responses, not Chat Completions). It carries a genuine long-context tier above 272K input tokens (input/cached-input/cache-write rates double, output rises 1.5x), the same shape as the GPT-5.6 family.

Best for
  • β€’ Frontier reasoning, coding, and agentic workloads that need OpenAI's latest model via the Responses API
Reach for something else if
  • β€’ Integrations that only support Chat Completions and cannot call the Responses API
Context Length
1.1M
Max Output
128K
Input Priceper 1M tokens
$10.80/ 1M tokens
Output Priceper 1M tokens
$54.00/ 1M tokens

Modalities

textimage→text

Capabilities

Vision

Pricing Breakdown

TypeRate
Prompts up to 272K tokens
Input$10.80 / 1M tokens
Output$54.00 / 1M tokens
Cached input$1.08 / 1M tokens
Cache write (5m)$13.50 / 1M tokens
Prompts over 272K tokens
Input$21.60 / 1M tokens
Output$81.00 / 1M tokens
Cached input$2.16 / 1M tokens
Cache write$27.00 / 1M tokens

Tiered pricing: when a request's prompt (input, cached tokens included) is over 272K tokens, every token in that request is billed at the long-context rates; exactly 272K is still the base rate.

Supported Parameters

max_tokenstoolstool_choiceresponse_formatstopreasoning

Specifications

Context window1,050,000 tokensverified
Max output tokens128,000 tokensverified
Release dateSeptember 3, 2026verified
LicenseCommercial via OpenAI APIverified

Benchmarks

BenchmarkDistributionScoreSource
Not publicly disclosed
β€”Not publicly disclosedβ€”

API Usage Examples

Use the global api.therouter.ai endpoint shown below for new integrations; the legacy China accelerated endpoint is retired.

cURL
curl https://api.therouter.ai/v1/chat/completions   -H "Content-Type: application/json"   -H "Authorization: Bearer $THE_ROUTER_API_KEY"   -d '{
    "model": "openai/gpt-6-astra",
    "messages": [
      {"role": "user", "content": "Summarize the key points from this input."}
    ]
  }'

API guide

Responses API

cURL
curl https://api.therouter.ai/v1/responses \
  -H "Authorization: Bearer $THEROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"openai/gpt-6-astra","input":"Hello"}'

More from openai

Similar models

Cross-provider sibling models

Related news

Guides

Fact ledger β€” every claim on this page traces here
sourceURLretrieved
Context windowβ€”β€”verified
Max output tokensβ€”β€”verified
Release dateβ€”β€”verified
Licenseβ€”β€”verified
Not publicly discloseddevelopers.openai.com β†—2026-09-16unknown
Help & contact