# Native model details: 16,000-token context and 2,000 max output tokens.
# https://developers.openai.com/api/docs/models/gpt-realtime-whisper
name = "GPT Realtime Whisper"
description = "Streaming speech-to-text model for low-latency transcript deltas from live audio"
family = "whisper"
release_date = "2026-05-07"
last_updated = "2026-05-07"
attachment = false
reasoning = false
temperature = true
tool_call = false
open_weights = false

[limit]
context = 16_000
output = 2_000

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