ChatGPT API Cost
The ChatGPT API cost is the per-use fee OpenAI charges developers and site owners to access GPT models programmatically — billed by token, not by the month.
What Is ChatGPT API Cost?
Who this is for: Small business owners or freelancers who want to add AI-powered features to a WordPress site and need to understand what OpenAI charges before committing.
Affiliate disclosure: WPSchool uses affiliate links to recommended products. This glossary entry is educational and contains no affiliate links.
The ChatGPT API cost is the per-use fee OpenAI charges developers and site owners to access GPT models programmatically — billed by token, not by the month.
Answer capsule: ChatGPT API cost refers to OpenAI’s usage-based pricing for accessing GPT models via API. As of April 2026, GPT-4.1 costs $2.00 per million input tokens and $8.00 per million output tokens. A typical chatbot reply of 500 words uses roughly 650 tokens, costing under $0.01.
What Is a Token?
A token is roughly 0.75 words of English text — “WordPress” counts as one token, “site” as one. Every request you send (input) and every reply you receive (output) is metered in tokens.
A 200-word product description you send plus a 300-word AI rewrite you receive equals roughly 650 tokens total. At GPT-4.1 rates, that exchange costs about $0.004 — less than half a cent.
How Does ChatGPT API Pricing Work?
OpenAI charges separately for input tokens (text you send) and output tokens (text the model returns). Output tokens cost more — typically 3–6× the input rate — because generating text is more compute-intensive than reading it.
Current published rates at OpenAI’s pricing page as of April 2026:
| Model | Input (per 1M tokens) | Output (per 1M tokens) |
|---|---|---|
| GPT-4.1 | $2.00 | $8.00 |
| GPT-4.1 mini | $0.40 | $1.60 |
| GPT-4o | $2.50 | $10.00 |
| GPT-3.5 Turbo | $0.50 | $1.50 |
For most WordPress use cases — chatbots, content helpers, form-response drafters — GPT-4.1 mini delivers near-GPT-4.1 quality at 80% lower cost. We default to it on client sites handling under 10,000 monthly interactions.
What Does This Mean for a WordPress Site?
A modest AI chatbot on a small business site handling 500 conversations per month (avg. 300 words in, 400 words out per conversation) consumes roughly 600,000 tokens. At GPT-4.1 mini rates, that is approximately $0.60/month — a cost most site owners will not notice.
Where costs escalate: high-volume content generation, image-generation endpoints (billed separately), and embedding large documents. In our testing with an ecommerce client running automated product descriptions across 2,000 SKUs, a single generation run cost $4.20 using GPT-4.1 mini.
Do You Need a Paid ChatGPT Subscription to Use the API?
No. The ChatGPT consumer subscription (ChatGPT Plus at $20/month) is a separate product from the API. API access requires an OpenAI platform account with billing enabled — you pay only for what you use, with no monthly minimum.
Related Terms
- Token — the unit OpenAI uses to measure text length for billing
- API key — the credential your WordPress plugin uses to authenticate with OpenAI
- Rate limit — the maximum requests-per-minute OpenAI allows on your account tier
- Prompt engineering — writing efficient inputs to reduce token consumption
- OpenAI Playground — a browser tool for testing API calls and estimating token usage before writing code
Additional Reading
- How to Add a Chatbot to WordPress Without Coding — plugin-based setup with no API configuration needed
- Best AI Plugins for WordPress in 2026 — tested options that handle API billing inside wp-admin
- WordPress Hosting for AI-Heavy Sites — why shared hosting stalls under API-driven traffic
Last verified: April 2026