GPT-6 Astra

Same model, served by 3 providers.

openaiClosed1.1M contextReleased 2026-09-04openai/gpt-6-astra

Providers

Provider	Their model id	Serving	Input	Output	Cached in	Blended 3:1	Source
Azure AI Foundry	`gpt-6-astra`	flexglobal endpointQuantization not disclosed	$5.0	$25.0	$0.50	$10.0	Provider API Verified 4h ago · Changed 7h ago More prices cache read (long context): $1.0 cache write: $6.25 cache write (long context): $12.5 input (long context): $10.0 output (long context): $37.5
Azure AI Foundry	`gpt-6-astra`	standardglobal endpointQuantization not disclosed	$10.0	$50.0	$1.0	$20.0	Provider API Verified 4h ago · Changed 7h ago More prices cache read (long context): $2.0 cache write: $12.5 cache write (long context): $25.0 input (long context): $20.0 output (long context): $75.0
OpenAI Original creator	`gpt-6-astra`	standardQuantization not disclosed	$10.0	$50.0	$1.0	$20.0	Unverified (via models.dev) Verified 6h ago · Changed 6h ago More prices cache read (prompts > 200k): $2.0 cache read (prompts > 272k): $2.0 cache write: $12.5 cache write (prompts > 200k): $25.0 cache write (prompts > 272k): $25.0 input (prompts > 200k): $20.0 input (prompts > 272k): $20.0 output (prompts > 200k): $75.0 output (prompts > 272k): $75.0
AWS Bedrock	`global.openai.gpt-6-astra`	standardQuantization not disclosed	$10.0	$50.0	$1.0	$20.0	Unverified (via models.dev) Verified 6h ago · Changed 6h ago More prices cache read (prompts > 200k): $2.0 cache read (prompts > 272k): $2.0 cache write: $12.5 cache write (prompts > 200k): $25.0 cache write (prompts > 272k): $25.0 input (prompts > 200k): $20.0 input (prompts > 272k): $20.0 output (prompts > 200k): $75.0 output (prompts > 272k): $75.0
Azure AI Foundry	`gpt-6-astra`	standarddata zone (US)Quantization not disclosed	$11.0	$55.0	$1.1	$22.0	Provider API Verified 4h ago · Changed 7h ago More prices cache read (long context): $2.2 cache write: $13.75 cache write (long context): $27.5 input (long context): $22.0 output (long context): $82.5
AWS Bedrock	`us.openai.gpt-6-astra`	standardQuantization not disclosed	$11.0	$55.0	$1.1	$22.0	Unverified (via models.dev) Verified 6h ago · Changed 6h ago More prices cache read (prompts > 200k): $2.2 cache read (prompts > 272k): $2.2 cache write: $13.75 cache write (prompts > 200k): $27.5 cache write (prompts > 272k): $27.5 input (prompts > 200k): $22.0 input (prompts > 272k): $22.0 output (prompts > 200k): $82.5 output (prompts > 272k): $82.5
AWS Bedrock	`openai.gpt-6-astra`	standardQuantization not disclosed	$11.0	$55.0	$1.1	$22.0	Unverified (via models.dev) Verified 6h ago · Changed 6h ago More prices cache read (prompts > 200k): $2.2 cache read (prompts > 272k): $2.2 cache write: $13.75 cache write (prompts > 200k): $27.5 cache write (prompts > 272k): $27.5 input (prompts > 200k): $22.0 input (prompts > 272k): $22.0 output (prompts > 200k): $82.5 output (prompts > 272k): $82.5

Same weights don't always mean identical outputs: quantization, context limits and serving stacks can change quality, especially for tool calling. Prices are USD per 1M tokens; hyperscaler prices are for the reference region shown.

Monthly cost estimate

Input tokens / month (M)Output tokens / month (M)Cache hit (% of input)% of requests over 200k tokens

Azure AI Foundry global endpoint	$2,000/mo
OpenAI	$2,000/mo
AWS Bedrock	$2,000/mo
Azure AI Foundry data zone (US)	$2,200/mo
AWS Bedrock	$2,200/mo
AWS Bedrock	$2,200/mo

Cache hits use each provider's cached-input price where it has one. Cache write costs aren't included.