gpt-4o-mini

GPT-4o mini is OpenAI's most cost-efficient, high-performance small model (SLM), offering multimodal (text and vision) capabilities at a fraction of the cost.

This is the 'mini' model that changes the scale game: GPT-4o mini delivers advanced intelligence and multimodal support (text, vision) at a dramatically reduced price point. It’s engineered for efficiency, costing only 15 cents per 1M input tokens—a 60%+ reduction over GPT-3.5 Turbo. Performance is solid: it scores 82% on MMLU and features a massive 128K context window, making it ideal for high-volume, low-latency applications like customer support chatbots or large-scale document analysis. Deploy this model when you need superior quality and speed without the frontier model expenditure.

https://openai.com/index/gpt-4o-mini-advancing-cost-efficient-intelligence/