GPT-4o mini Audio Preview

GPT-4o mini Audio Preview was OpenAI's cost-efficient audio-capable model, launched December 17, 2024 alongside the GPT-4o audio family. It accepted audio or text input and could produce audio or text output through the Chat Completions and Realtime APIs, targeting asynchronous voice workloads like recorded sentiment analysis and text-to-audio content creation at roughly a quarter of the cost of the full GPT-4o audio models. It carried a 128K context window with 16K max output tokens.

OpenAI's model documentation now lists gpt-4o-mini-audio-preview and its dated snapshot gpt-4o-mini-audio-preview-2024-12-17 as deprecated, superseded by newer audio models in the lineup. Pricing at launch was $0.15 per million text input tokens and $0.60 per million text output tokens, with audio tokens at $10 input and $20 output per million. It remains a useful historical reference for the generation of voice products built on the 2024-2025 OpenAI audio stack.