Overview
GPT-5.4 Nano is the most affordable model in the GPT-5.4 family, designed for high-volume, low-cost workloads where minimal latency and maximum throughput matter.
Key Features
- OpenAI-compatible API - Use
/vendors/openai/v1/chat/completions or /vendors/openai/v1/responses
- Ultra-low cost - The cheapest model in the GPT-5.4 lineup
- Cached input support - Further cost savings on repeated prompts
- Streaming support - Real-time token streaming
Supported Endpoints
| Endpoint | Path |
|---|
| Chat Completions | /vendors/openai/v1/chat/completions |
| Responses | /vendors/openai/v1/responses |
Usage
Set model to gpt-5.4-nano in your request body.
Pricing
Per Million Tokens
Input: $0.20
Cached Input: $0.02
Output: $1.25