Skip to content

GPT-6 Astra (batch)

file + image + text → textOpenRouter Cost calculator →

Best price · blended / 1M

$10

at OpenAI

Input / output
$5.00 / $25
Providers
4
Spread
8.0×
7 days
0%
30 days
0%
Context
1.1M
Prices per 1M tokens, blended: OpenAI (flex) $10, Amazon Bedrock (global) $20, Azure $20, Google (global) $20, OpenAI $20, OpenAI (fast) $40, OpenAI (ultrafast) $120.

Providers

  • OpenAICheapest
    flex
    $10
    $5.00 in · $25 out
    −33% vs ref
    1.1M ctx100% uptime
  • $20
    $10 in · $50 out
    +33% vs ref
    1.1M ctx100% uptime
  • $20
    $10 in · $50 out
    +33% vs ref
    1.1M ctx100% uptime
  • global
    $20
    $10 in · $50 out
    +33% vs ref
    1.1M ctx
  • $20
    $10 in · $50 out
    +33% vs ref
    1.1M ctx99.7% uptime
  • fast
    $40
    $20 in · $100 out
    +167% vs ref
    1.1M ctx100% uptime
  • ultrafast
    $120
    $60 in · $300 out
    +700% vs ref
    1.1M ctx
Blended = (3 × input + output) / 4. Ref $15 = median of cheapest 20%.

Price history

USD per 1M tokens · each step is a price change

Prices in USD / 1M tokens.

No provider has changed its price since tracking began on Oct 3, 2026.

Long-context pricing

Offered byPrompt sizeInputOutput
OpenAI (flex)base$5.00$25
> 272K$10$37.50
Amazon Bedrock (global), Azure +2base$10$50
> 272K$20$75
OpenAI (fast)base$20$100
> 272K$40$150
OpenAI (ultrafast)base$60$300
> 272K$120$450

Other pricing

Offered byCache readCache writeWeb search
OpenAI (flex)$0.50$6.25$0.01
Amazon Bedrock (global), Azure +2$1.00$12.50$0.01
OpenAI (fast)$2.00$25$0.01
OpenAI (ultrafast)$6.00$75$0.01

Changes

  • Now served,
    Listed
  • No longer served,
    Delisted
  • No longer served,
    Delisted
  • Now served,
    Listed

About

The cheapest provider for GPT-6 Astra (batch) is OpenAI at $10 per 1M tokens (blended), charging $5.00 per 1M input and $25 per 1M output tokens. 4 providers serve it; the most expensive charges 8.0× the typical price. Context window: 1.1M tokens. Prices tracked since Oct 3, 2026.

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...
Released
Sep 4, 2026
First tracked
Oct 3, 2026
Canonical slug
openai/gpt-6-astra-20260903