GPT-5.4 Nano API Pricing & Cost Calculator

Q: Does GPT-5.4 Nano support prompt caching?

Yes, OpenAI offers a 50% discount on cached input tokens, reducing the input price to $0.075 per million for matches.

GPT-5.4 Nano is OpenAI's most cost-efficient reasoning model, delivering strong performance at a fraction of the flagship price. It is ideal for high-volume classification, extraction, and simple agent tasks.

Input tokens

50,000

50% cached

Output tokens

5,000

per request

Volume

1,000 / monthly

Standard API

Cost Comparison

Based on 50,000 input tokens (50% cached), 5,000 output tokens, and 1,000 requests.

GPT-5.4 Nano

$11.75

Gemini 3.5 Flash

$33.75

Claude Haiku 4.5

$52.50

Gemini 3.1 Pro (<=200k)

$115.00

GPT-5.4

$143.75

Claude Sonnet 4.6

$157.50

Claude Opus 4.7

$262.50

GPT-5.5

$287.50

Open full calculator to adjust tokens, caching, or add other models

How this is calculated

GPT-5.4 Nano is priced at $0.15 per million input tokens and $0.60 per million output tokens, supporting a 50% caching discount ($0.075 per million) and a 50% batch discount ($0.075 per million inputs, $0.30 per million outputs).

Verdict

GPT-5.4 Nano is the go-to choice for cost-sensitive production workloads. Prompt caching and batch processing push costs down to near-zero for repetitive inference tasks.