GPT-5.6 Luna

Same model, served by 3 providers.

openaiClosed1.1M contextReleased 2026-07-09openai/gpt-5.6-luna

Providers

Show flex / batch / priority tiers Full precision only

Provider	Their model id	Serving	Input	Output	Cached in	Blended 3:1	Source
Azure AI Foundry	`gpt-5.6-luna`	standardglobal endpointQuantization not disclosed	$0.20	$1.2	$0.02	$0.45	Provider API Verified 3h ago · Changed 6h ago More prices cache read (long context): $0.04 cache write: $0.25 cache write (long context): $0.50 input (long context): $0.40 output (long context): $1.8
OpenAI Original creator	`gpt-5.6-luna`	standardQuantization not disclosed	$0.20	$1.2	$0.02	$0.45	Unverified (via models.dev) Verified 6h ago · Changed 6h ago More prices cache read (prompts > 200k): $0.04 cache read (prompts > 272k): $0.04 cache write: $0.25 cache write (prompts > 200k): $0.50 cache write (prompts > 272k): $0.50 input (prompts > 200k): $0.40 input (prompts > 272k): $0.40 output (prompts > 200k): $1.8 output (prompts > 272k): $1.8
AWS Bedrock	`global.openai.gpt-5.6-luna`	standardQuantization not disclosed	$0.20	$1.2	$0.02	$0.45	Unverified (via models.dev) Verified 6h ago · Changed 6h ago More prices cache read (prompts > 200k): $0.04 cache read (prompts > 272k): $0.04 cache write: $0.25 cache write (prompts > 200k): $0.50 cache write (prompts > 272k): $0.50 input (prompts > 200k): $0.40 input (prompts > 272k): $0.40 output (prompts > 200k): $1.8 output (prompts > 272k): $1.8
Azure AI Foundry	`gpt-5.6-luna`	standarddata zone (US)Quantization not disclosed	$0.22	$1.32	$0.022	$0.495	Provider API Verified 3h ago · Changed 6h ago More prices cache read (long context): $0.044 cache write: $0.275 cache write (long context): $0.55 input (long context): $0.44 output (long context): $1.98
AWS Bedrock	`openai.gpt-5.6-luna`	standardQuantization not disclosed	$0.22	$1.32	$0.022	$0.495	Unverified (via models.dev) Verified 6h ago · Changed 6h ago More prices cache read (prompts > 200k): $0.044 cache read (prompts > 272k): $0.044 cache write: $0.275 cache write (prompts > 200k): $0.55 cache write (prompts > 272k): $0.55 input (prompts > 200k): $0.44 input (prompts > 272k): $0.44 output (prompts > 200k): $1.98 output (prompts > 272k): $1.98
AWS Bedrock	`us.openai.gpt-5.6-luna`	standardQuantization not disclosed	$0.22	$1.32	$0.022	$0.495	Unverified (via models.dev) Verified 6h ago · Changed 6h ago More prices cache read (prompts > 200k): $0.044 cache read (prompts > 272k): $0.044 cache write: $0.275 cache write (prompts > 200k): $0.55 cache write (prompts > 272k): $0.55 input (prompts > 200k): $0.44 input (prompts > 272k): $0.44 output (prompts > 200k): $1.98 output (prompts > 272k): $1.98
AWS Bedrock	`in.openai.gpt-5.6-luna`	standardQuantization not disclosed	$0.22	$1.32	$0.022	$0.495	Unverified (via models.dev) Verified 6h ago · Changed 6h ago More prices cache read (prompts > 200k): $0.044 cache read (prompts > 272k): $0.044 cache write: $0.275 cache write (prompts > 200k): $0.55 cache write (prompts > 272k): $0.55 input (prompts > 200k): $0.44 input (prompts > 272k): $0.44 output (prompts > 200k): $1.98 output (prompts > 272k): $1.98

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	$44/mo
OpenAI	$44/mo
AWS Bedrock	$44/mo
Azure AI Foundry data zone (US)	$48.4/mo
AWS Bedrock	$48.4/mo
AWS Bedrock	$48.4/mo
AWS Bedrock	$48.4/mo

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