GPT-5 Pricing and Offerings Update

Pricing | OpenAI
Try ChatGPT     (opens in a new window)
Try ChatGPT     (opens in a new window)
Our most capable model for professional work
Our strongest mini model yet for coding, computer use, and subagents
Our cheapest GPT-5.4-class model for simple high-volume tasks
Pricing above reflects standard processing rates for context lengths under 270K. See the full pricing page    here ⁠ (opens in a new window)   .    Data residency and Regional Processing ⁠ (opens in a new window)    endpoints are charged an additional 10% for all GPT 5.4 models.
To optimize cost and performance for different use cases, we also offer:
Batch API⁠   ⁠ (opens in a new window)   : Save 50% on inputs and outputs with the Batch API and run tasks asynchronously over 24 hours.
Priority processing     ⁠  ⁠  : offers reliable, high-speed performance with the flexibility to pay-as-you-go.
Our APIs
Realtime API

API Pricing | OpenAI
Log in      Try ChatGPT     (opens in a new window)
Try ChatGPT     (opens in a new window)   Login
Our most capable model for professional work.
Our strongest mini model yet for coding, computer use, and subagents.
Our cheapest GPT-5.4-class model for simple high-volume tasks.
Pricing above reflects standard processing rates for context lengths under 270K.
Data residency and Regional Processing   ⁠ (opens in a new window)    ⁠ endpoints are charged an additional 10% for all models released after 3/5/26.
Multimodal models
Power applications across text, image, audio, and video with models built for real-time interaction and rich media generation.
GPT-realtime-1.5
Our most capable model for realtime voice interactions.