OpenAI reveals the pricing plans for its API — and it isn’t cheap

OpenAI has revealed the projected pricing plans for its API, which lets individuals use the corporate’s vaunted AI tools on “virtually any English language task.” But you’re gonna need money to burn in case you wanna try it out.

The API gives customers access to GPT-3, OpenAI‘s headline-grabbing text generator. The firm says developers can apply it “to any language task — semantic search, summarization, sentiment analysis, content generation, translation, and more — with only a few examples or by specifying your task in English.”

The product was initially launched in a free, two-month non-public beta on July 11. But from October 1, users of the beta model will have to choose between 4 different pricing plans to access the system. A researcher called Gwern Branwen posted the details on Reddit

1. Explore: Free tier: 100K tokens or a three-month trial, whichever you use up first.
2. Create: $100 monthly for 2M tokens, plus 8 cents for every additional 1k tokens.
3. Build: $400 monthly for 10M tokens, plus 6 cents for every additional 1k tokens.
4. Scale: Contact OpenAI for pricing.

According to Branwen, 2 million tokens is roughly equal to 3,000 pages of text. For reference, the entire works of Shakespeare contain roughly 900,000 words.

The prices are subject to change, however the preliminary pricing will put the model out of reach for most hobbyists. However, they are going to get three months to experiment with the system without spending a dime.

Jay Alammar, a machine learning instructor and VC, noted that OpenAI‘s lead in the area has given it some room to charge what it likes.

“Competitors will come (with free tiers and lower prices) eventually,” he mentioned on Twitter.