name = "OpenAI: GPT Audio"
description = "Speech generation model for controllable voice, narration, and audio delivery"
release_date = "2026-01-20"
last_updated = "2026-03-15"
attachment = false
reasoning = false
temperature = true
tool_call = false
open_weights = false

[cost]
input = 2.5
output = 10

[limit]
context = 128000
output = 16384

[modalities]
input = ["audio", "text"]
output = ["audio", "text"]