name = "GPT-Realtime-2.1" description = "Realtime speech-to-speech model with configurable reasoning, tool use, and robust voice-agent behavior" family = "gpt" release_date = "2026-07-06" last_updated = "2026-07-06" attachment = true reasoning = true temperature = false tool_call = true structured_output = false knowledge = "2024-09-30" open_weights = false [limit] context = 128_000 input = 96_000 output = 32_000 [modalities] input = ["text", "audio", "image"] output = ["text", "audio"]