Skip to content

gpt-5-mini

Retirement scheduled gpt-5-mini Retires Jul 13, 2027 · in 8 months
Replacement Not named yet Microsoft usually names one before retirement. How retirement works
Broad GA · until Jul 2027 OpenAI 25+ regions tracked

Available in 29 of 35 tracked regions with 6 deployment SKU types.

Datazone Global Provisioned
Total regions 29
Coverage 83%
SKU types 6
Next retirement in 8 months
Widest SKU footprint Provisioned global 29 regions · 83% coverage · Provisioned

Lifecycle

2025-08-07 Generally available in 8 months
Released Aug 7, 2025Retires Jul 13, 2027

Pricing

Pay-as-you-go · Global $0.6875 per 1M tokens Blended at 3:1 input:output
Project monthly cost
Per 1M tokensGlobalData Zone
Input$0.25$0.275
Cached input$0.025$0.0275
Output$2.00$2.20
Batch input$0.125$0.1375
Batch output$1.00$1.10

Provisioned (PTU): Global $1.00 · Data Zone $1.10 · Regional $2.00 per PTU-hour. A 1-year reservation brings it to about $221 per PTU-month. Minimum 15 PTUs (Global / Data Zone), about 4,750 input tokens per minute each.

Pay-as-you-go list price in USD for East US 2 from the Azure Retail Prices API, checked daily, last updated 2026-11-01. Excludes tax and negotiated discounts; other regions can differ.

Deployment Options

Global

Global deployments

Worldwide availability with intelligent routing

Best for applications needing global reach with automatic failover

Global Standard 26 regions · 74% coverage
Global coverage 29 regions · 83% coverage

⚠ Data may be processed in any Azure region — not suitable for HIPAA, FedRAMP, or strict data-residency requirements

Datazone

Datazone deployments

Data residency compliance deployments

Required for data sovereignty and compliance requirements (GDPR, etc.)

Datazone provisioned managed 13 regions · 37% coverage
Datazone standard 14 regions · 40% coverage

✓ Data stays within the specified geographic zone — supports GDPR and regional data-residency policies

Provisioned

Provisioned deployments

Reserved throughput capacity (PTU)

Best for predictable, high-volume production workloads

Provisioned (PTU managed) 10 regions · 29% coverage
Provisioned global 29 regions · 83% coverage

✓ Single-region deployment — HIPAA-eligible in supported regions with a BAA from Microsoft

Full Availability Matrix

29 regions
RegionDatazone provisioned managedDatazone standardGlobal StandardGlobal coverageProvisioned (PTU managed)Provisioned global
Australia East——✓✓✓✓
Brazil South——✓✓—✓
Canada Central———✓—✓
Canada East——✓✓✓✓
Central US—✓✓✓—✓
East US✓✓✓✓—✓
East US 2✓✓✓✓✓✓
France Central✓✓✓✓—✓
Germany West Central✓✓✓✓—✓
Italy North✓✓✓✓—✓
Japan East——✓✓✓✓
Korea Central——✓✓✓✓
North Central US✓✓✓✓—✓
North Europe———✓—✓
Norway East——✓✓—✓
Poland Central✓✓✓✓—✓
South Africa North——✓✓—✓
South Central US✓✓✓✓—✓
South India——✓✓✓✓
Southeast Asia——✓✓—✓
Spain Central✓✓✓✓—✓
Sweden Central✓✓✓✓✓✓
Switzerland North——✓✓—✓
Switzerland West———✓—✓
UAE North——✓✓✓✓
UK South——✓✓✓✓
West Europe✓✓✓✓—✓
West US✓✓✓✓—✓
West US 3✓✓✓✓✓✓

← Back to All Models

Last updated: 2026-11-12 16:36 UTC