Skip to main content
⚡ EfficientVery Cheap

GPT-5 nano

by OpenAI

Ultra-fast and cheapest OpenAI model for simple tasks

Try OpenAI →

Input Price

$0.05

per 1M tokens

Output Price

$0.4

per 1M tokens

Context Window

128K

tokens (8.192K max output)

Specifications

Release Date2025-08-07
Category⚡ Efficient
Context Window128,000 tokens
Max Output8,192 tokens
ProviderOpenAI
Input Cost$0.05 / 1M tokens
Output Cost$0.4 / 1M tokens
Price TierVery Cheap

Capabilities

text

Related Use Cases

💬

AI Chatbot

Calculate the API cost of running a customer support or conversational AI chatbot.

~2,000 in · ~500 out
📄

Document Summarizer

Calculate how much it costs to summarize documents, reports, and articles with AI.

~8,000 in · ~1,000 out
✍️

Content Generation

Calculate the cost of generating blog posts, marketing copy, and articles with AI.

~1,500 in · ~3,000 out
🔍

Data Extraction

Calculate the cost of extracting structured data from unstructured text with AI.

~3,000 in · ~1,500 out
🌐

AI Translation

Calculate the cost of translating text with AI language models.

~2,000 in · ~2,500 out
📧

Email Assistant

Calculate the cost of an AI-powered email drafting and response assistant.

~1,500 in · ~800 out

Monthly Cost Estimates

Usage LevelDaily TokensDailyMonthlyYearly
Light~50 requests/day100K in / 20K out$0.01$0.39$5
Medium~200 requests/day500K in / 100K out$0.07$1.95$24
Heavy~1K requests/day2,000K in / 500K out$0.30$9.00$110
Enterprise~5K requests/day10,000K in / 2,000K out$1.30$39.00$475

Cost Calculator

Alternatives to GPT-5 nano

💰 Cheaper Options

DeepSeek V4 FlashDeepSeek
7% cheaper
$0.14/M in · $0.28/M out
Ministral 3 14BMistral AI
11% cheaper
$0.2/M in · $0.2/M out
Mistral Small 3.2Mistral AI
11% cheaper
$0.1/M in · $0.3/M out
Gemini Embedding 2Google
11% cheaper
$0.2/M in · $0.2/M out

🚀 More Powerful

GPT-4.1 nanoOpenAI
11% more
$0.1/M in · $0.4/M out
Gemini 2.0 FlashGoogle
11% more
$0.1/M in · $0.4/M out
Gemini 2.5 Flash-LiteGoogle
11% more
$0.1/M in · $0.4/M out
Voxtral SmallMistral AI
11% more
$0.1/M in · $0.4/M out

Frequently Asked Questions

How much does GPT-5 nano API cost per million tokens?▼
GPT-5 nano costs $0.05 per million input tokens and $0.4 per million output tokens as of 2026. These are the standard API rates from OpenAI.
What is the GPT-5 nano context window?▼
GPT-5 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 nano cost per month?▼
At medium usage (~200 requests/day with 500K input and 100K output tokens/day), GPT-5 nano costs approximately $1.95/month. Light usage runs about $0.39/month, and heavy usage (~1K requests/day) around $9.00/month.
Is there a cheaper alternative to GPT-5 nano?▼
Yes — DeepSeek V4 Flash by DeepSeek is a cheaper option at $0.14/M input tokens vs $0.05/M for GPT-5 nano. Other budget alternatives include models in the efficient tier.
Is GPT-5 nano good for ai chatbot?▼
GPT-5 nano is a efficient model with support for text. For ai chatbot, it offers 128K context and costs $0.05/M input tokens — a budget-friendly choice for this use case.

GPT-5 nano Comparisons

codex mini latest vs gpt 5 nanogpt 5 nano vs o4 mini deep researchgpt 5 nano vs o3 deep researchgpt 5 2 vs gpt 5 nanogpt 5 2 pro vs gpt 5 nanogpt 5 6 sol vs gpt 5 nanogpt 5 6 terra vs gpt 5 nanogpt 5 6 luna vs gpt 5 nano

Ready to use GPT-5 nano?

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

Try OpenAI API →

Related Articles

GPT Live Voice Workflows: What OpenAI's August 2026 Update Makes Possible

GPT Live adds full-duplex voice, async GPT-5.5 delegation, and faster startup. Here are 7 workflows teams can build now.

8/4/2026

Did the Codex Default Model Change to GPT-5.4-Codex-Mini?

If you saw searches or changelog chatter about a Codex default model change to GPT-5.4-Codex-Mini, here is the short...

8/3/2026

GPT-5.6 Price-Performance Frontier: Workflows Builders Can Ship Now

GPT-5.6 brings Sol, Terra, and Luna tiers for faster agents, document synthesis, code review, support ops, and smarte...

7/31/2026

← Browse all providers & models