← All models

GPT-Audio mini

Published Publisher OpenAIgpt-audio-mini

Overview

GPT-Audio mini is a cost-efficient GPT Audio model for speech-to-speech applications. It accepts and produces audio and text, and is documented for use through the Chat Completions API. Potential uses include voice-driven assistants and spoken conversations. (developers.openai.com)

Release and availability

OpenAI announced GPT-Audio mini on October 6, 2025. The associated snapshot was gpt-audio-mini-2025-10-06. That snapshot was shut down on July 23, 2026; OpenAI lists gpt-audio-1.5 as a substitute. The gpt-audio-mini alias is distinct from the retired snapshot and was updated to point to the December 15, 2025 snapshot. (developers.openai.com)

Capabilities and limitations

The current model page lists a 128,000-token context window, a maximum of 16,384 output tokens, and an October 1, 2023 knowledge cutoff. It lists function calling as supported, but streaming, structured outputs, fine-tuning, image input, and video input as unsupported. These are model-page specifications and are not independently verified for the retired October snapshot. (developers.openai.com)

Pricing

As checked on November 17, 2026, the official model page lists USD $0.60 per 1 million input text tokens and USD $2.40 per 1 million output text tokens for gpt-audio-mini. These are model-page prices, not a verified historical quote for the shut-down October snapshot. Audio input and output prices were not verified. (developers.openai.com)

Official sources