OpenAI has launched GPT-4o mini, a smaller and more cost-efficient AI model than its current models. GPT-4o mini is available for developers and through ChatGPT for Free, plus, and Team users, replacing GPT-3.5. Enterprise users will gain access next week.

GPT-4o mini reportedly outperforms other small AI models on reasoning tasks involving text and vision. It consists of text and vision capabilities, with support for video and audio planned for the future. The model has a context window of 128,000 tokens, roughly the length of a book, and a knowledge cutoff of October 2023. OpenAI also claims that the GPT-4o mini is significantly more affordable to run than its previous models and more than 60% cheaper than the GPT-3.5 Turbo.

Analyst QuickTake: With this, OpenAI joins the ranks of other companies such as Microsoft ( Phi-3 Mini ), Google (Gemini Nano), Stability AI (Stable LM 2 1.6B), and Apple ( OpenELM ) which has previously launched small language models (SLMs). These models have already seen use cases, for instance, Google’s Gemini Nano which can operate locally on smartphones was integrated with MediaTek’s Dimensity 9300 and 8300 chips to enable AI features on mid-range phones.