ChatGPT 4o Mini

GPT-4o Mini delivers advanced, multimodal intelligence (text and vision) at an unprecedented cost: 15 cents per million input tokens.

This is OpenAI's most cost-efficient small model, replacing GPT-3.5 Turbo as the new baseline for high-volume AI tasks. It offers a 128K token context window and robust multimodal capabilities (text and vision), all with low latency. Benchmark results confirm its power: 82.0% on MMLU and 87.0% on MGSM, outperforming competitors like Gemini Flash and Claude Haiku. The aggressive API pricing—15 cents per million input tokens—makes advanced AI highly accessible for real-time applications and customer support.

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