Skip to main content
⚡ EfficientVery Cheap

GPT-5.4 nano

by OpenAI

Ultra-fast and cheapest GPT-5.4 variant for simple tasks

Try OpenAI

Input Price

$0.2

per 1M tokens

Output Price

$1.25

per 1M tokens

Context Window

128K

tokens (8.192K max output)

Specifications

Release Date2026-03-06
Category⚡ Efficient
Context Window128,000 tokens
Max Output8,192 tokens
ProviderOpenAI
Input Cost$0.2 / 1M tokens
Output Cost$1.25 / 1M tokens
Price TierVery Cheap

Capabilities

text

Related Use Cases

Monthly Cost Estimates

Usage LevelDaily TokensDailyMonthlyYearly
Light~50 requests/day100K in / 20K out$0.05$1.35$16
Medium~200 requests/day500K in / 100K out$0.23$6.75$82
Heavy~1K requests/day2,000K in / 500K out$1.02$30.75$374
Enterprise~5K requests/day10,000K in / 2,000K out$4.50$135.00$1643

Cost Calculator

Alternatives to GPT-5.4 nano

Frequently Asked Questions

How much does GPT-5.4 nano API cost per million tokens?
GPT-5.4 nano costs $0.2 per million input tokens and $1.25 per million output tokens as of 2026. These are the standard API rates from OpenAI.
What is the GPT-5.4 nano context window?
GPT-5.4 nano supports a 128K context window (128,000 tokens), which means you can process up to 128K tokens in a single API call.
How much does GPT-5.4 nano cost per month?
At medium usage (~200 requests/day with 500K input and 100K output tokens/day), GPT-5.4 nano costs approximately $6.75/month. Light usage runs about $1.35/month, and heavy usage (~1K requests/day) around $30.75/month.
Is there a cheaper alternative to GPT-5.4 nano?
Yes — Codestral by Mistral AI is a cheaper option at $0.3/M input tokens vs $0.2/M for GPT-5.4 nano. Other budget alternatives include models in the efficient tier.
Is GPT-5.4 nano good for ai chatbot?
GPT-5.4 nano is a efficient model with support for text. For ai chatbot, it offers 128K context and costs $0.2/M input tokens — a budget-friendly choice for this use case.

GPT-5.4 nano Comparisons

Ready to use GPT-5.4 nano?

Get started with OpenAI's API — free tier available for most models.

Try OpenAI API →