OpenAI News

OpenAI Releases GPT-4o Mini — Fast and Affordable AI for Everyone

By · · Source: OpenAI · 3 min read
📰 Read Full Article at Source →

What Is GPT-4o Mini?

OpenAI has launched GPT-4o Mini, its most cost-efficient small model to date. Designed for developers and businesses that need high-volume AI processing at low cost, GPT-4o Mini delivers strong performance at a fraction of the price of larger models.

Performance and Benchmarks

GPT-4o Mini scores 82% on the MMLU benchmark, outperforming GPT-3.5 Turbo on most standard evaluations including reasoning, coding, and language understanding. The model supports both text and vision inputs, making it suitable for a wide range of applications from customer service to document analysis.

Pricing — 60% Cheaper Than GPT-3.5

GPT-4o Mini is priced at $0.15 per million input tokens and $0.60 per million output tokens — approximately 60% cheaper than GPT-3.5 Turbo. For businesses running millions of API calls per month, this represents a significant cost reduction without sacrificing quality on most tasks.

Availability

GPT-4o Mini is available immediately via the OpenAI API and is also being rolled out to ChatGPT Free tier users, replacing GPT-3.5 Turbo as the default free model. Developers can access it using the model string gpt-4o-mini in their API calls.

Use Cases

The model is ideal for applications requiring fast, affordable AI processing at scale: customer support chatbots, content moderation, data extraction, summarization, and classification tasks. Its vision capability also makes it useful for image understanding at scale.

Impact on the AI Market

The launch puts pressure on competitors including Anthropic, Google, and Meta who offer similarly positioned small models. GPT-4o Mini directly competes with Claude 3 Haiku and Gemini 1.5 Flash in the cost-efficient AI segment that is growing rapidly as enterprise AI adoption increases.