{"models":[{"id":"openai/gpt-realtime","type":"realtime","name":"GPT Realtime","model_id":"gpt-realtime","capabilities":["audio","text","tools","vad"],"speed":"fast","cost":0.457,"quality":"best","context_window":32000,"max_output_tokens":4096,"description":"Model capable of realtime text and audio inputs and outputs.","params":{"realtime":{"voices":["alloy","ash","ballad","coral","echo","sage","shimmer","verse"],"default_voice":"alloy","input_audio_format":"pcm16","output_audio_format":"pcm16","sample_rate":24000,"supports_vad":true,"supports_tools":true}},"pricing":{"billing_unit":"per_mtok","input":[4],"output":[24],"cached_input":[0.4],"audio_input":[32],"audio_output":[64],"audio_cached_input":[0.4],"image_input":[5],"image_cached_input":[0.5],"input_transcription_per_minute":{"gpt-4o-mini-transcribe":0.003,"gpt-4o-transcribe":0.006,"gpt-realtime-whisper":0.017,"whisper-1":0.006}},"region":"US","model":"gpt-realtime","family":"gpt-realtime","pooled":true,"provider":{"slug":"openai","display_name":"OpenAI","owner_country":"US"},"service_route":{"id":"openai/api-us","adapter":"openai"},"maker":{"slug":"openai","display_name":"OpenAI","country":"US"},"country":"United States","compliance":{"residency":"US","country":"United States","inference_location":"US","zdr":{"logging":true,"moderation":true,"caching":true,"training":false,"subprocessors":null},"training":"no","logging":"abuse_monitoring","retention_days":30,"moderation":"default_on","caching":"implicit","cache_scope":"organization","cache_ttl_seconds":1800,"content_storage":"retained","dpa_available":true,"third_party_access":true,"transfer_mechanism":"sccs","route_id":"openai/api-us"},"references":[{"kind":"model_card","url":"https://developers.openai.com/api/docs/models/gpt-realtime","created_at":"2026-10-07T05:57:44.064854Z","updated_at":"2026-10-07T05:57:44.064854Z"}],"input_modalities":["text","image","audio"],"output_modalities":["text","audio"]},{"id":"openai/gpt-realtime-1.5","type":"realtime","name":"GPT Realtime 1.5","model_id":"gpt-realtime-1.5","capabilities":["audio","text","vision","tools","vad"],"speed":"fast","cost":0.404,"quality":"best","context_window":32000,"max_output_tokens":4096,"description":"The best voice model for audio in, audio out. Real-time multimodal conversations with audio, text, and vision.","params":{"realtime":{"voices":["alloy","ash","ballad","coral","echo","fable","onyx","nova","sage","shimmer","verse"],"default_voice":"alloy","input_audio_format":"pcm16","output_audio_format":"pcm16","sample_rate":24000,"supports_vad":true,"supports_tools":true}},"pricing":{"billing_unit":"per_mtok","input":[4],"output":[16],"cached_input":[0.4],"audio_input":[32],"audio_output":[64],"audio_cached_input":[0.4],"image_input":[5],"image_cached_input":[0.5],"input_transcription_per_minute":{"gpt-4o-mini-transcribe":0.003,"gpt-4o-transcribe":0.006,"gpt-realtime-whisper":0.017,"whisper-1":0.006}},"region":"US","model":"gpt-realtime-1.5","family":"gpt-realtime-1.5","pooled":true,"provider":{"slug":"openai","display_name":"OpenAI","owner_country":"US"},"service_route":{"id":"openai/api-us","adapter":"openai"},"maker":{"slug":"openai","display_name":"OpenAI","country":"US"},"country":"United States","compliance":{"residency":"US","country":"United States","inference_location":"US","zdr":{"logging":true,"moderation":true,"caching":true,"training":false,"subprocessors":null},"training":"no","logging":"abuse_monitoring","retention_days":30,"moderation":"default_on","caching":"implicit","cache_scope":"organization","cache_ttl_seconds":1800,"content_storage":"retained","dpa_available":true,"third_party_access":true,"transfer_mechanism":"sccs","route_id":"openai/api-us"},"release":{"date":"2026-02-23","source":"vendor","url":"https://developers.openai.com/api/docs/changelog"},"references":[{"kind":"model_card","url":"https://developers.openai.com/api/docs/models/gpt-realtime-1.5","created_at":"2026-10-07T05:57:44.064854Z","updated_at":"2026-10-07T05:57:44.064854Z"}],"input_modalities":["text","image","audio"],"output_modalities":["text","audio"]},{"id":"openai/gpt-realtime-2","type":"realtime","name":"GPT Realtime 2","model_id":"gpt-realtime-2","capabilities":["audio","text","vision","tools","vad","reasoning"],"speed":"fast","cost":0.457,"quality":"best","context_window":128000,"max_output_tokens":32000,"description":"OpenAI's most capable realtime voice model. Audio, text, and image input with audio and text output, configurable reasoning effort, and reliable tool use.","params":{"realtime":{"voices":["alloy","ash","ballad","coral","echo","sage","shimmer","verse","marin","cedar"],"default_voice":"marin","input_audio_format":"pcm16","output_audio_format":"pcm16","sample_rate":24000,"supports_vad":true,"supports_tools":true,"reasoning":{"supported":["minimal","low","medium","high","xhigh"],"default":"low"}}},"pricing":{"billing_unit":"per_mtok","input":[4],"output":[24],"cached_input":[0.4],"audio_input":[32],"audio_output":[64],"audio_cached_input":[0.4],"image_input":[5],"image_cached_input":[0.5],"input_transcription_per_minute":{"gpt-4o-mini-transcribe":0.003,"gpt-4o-transcribe":0.006,"gpt-realtime-whisper":0.017,"whisper-1":0.006}},"featured":true,"region":"US","model":"gpt-realtime-2","family":"gpt-realtime-2","pooled":true,"provider":{"slug":"openai","display_name":"OpenAI","owner_country":"US"},"service_route":{"id":"openai/api-us","adapter":"openai"},"maker":{"slug":"openai","display_name":"OpenAI","country":"US"},"country":"United States","compliance":{"residency":"US","country":"United States","inference_location":"US","zdr":{"logging":true,"moderation":true,"caching":true,"training":false,"subprocessors":null},"training":"no","logging":"abuse_monitoring","retention_days":30,"moderation":"default_on","caching":"implicit","cache_scope":"organization","cache_ttl_seconds":1800,"content_storage":"retained","dpa_available":true,"third_party_access":true,"transfer_mechanism":"sccs","route_id":"openai/api-us"},"release":{"date":"2026-05-07","source":"vendor","url":"https://developers.openai.com/api/docs/changelog"},"references":[{"kind":"model_card","url":"https://developers.openai.com/api/docs/models/gpt-realtime-2","created_at":"2026-10-07T05:57:44.064854Z","updated_at":"2026-10-07T05:57:44.064854Z"}],"input_modalities":["text","image","audio"],"output_modalities":["text","audio"]},{"id":"openai/gpt-realtime-mini","type":"realtime","name":"GPT Realtime Mini","model_id":"gpt-realtime-mini","capabilities":["audio","text","vision","tools","vad"],"speed":"fast","cost":0.154,"quality":"great","context_window":32000,"max_output_tokens":4096,"description":"A cost-efficient version of GPT Realtime for audio and text conversations.","params":{"realtime":{"voices":["alloy","ash","ballad","coral","echo","sage","shimmer","verse"],"default_voice":"alloy","input_audio_format":"pcm16","output_audio_format":"pcm16","sample_rate":24000,"supports_vad":true,"supports_tools":true}},"pricing":{"billing_unit":"per_mtok","input":[0.6],"output":[2.4],"cached_input":[0.06],"audio_input":[10],"audio_output":[20],"audio_cached_input":[0.3],"image_input":[0.8],"image_cached_input":[0.08],"input_transcription_per_minute":{"gpt-4o-mini-transcribe":0.003,"gpt-4o-transcribe":0.006,"gpt-realtime-whisper":0.017,"whisper-1":0.006}},"region":"US","model":"gpt-realtime-mini","family":"gpt-realtime-mini","pooled":true,"provider":{"slug":"openai","display_name":"OpenAI","owner_country":"US"},"service_route":{"id":"openai/api-us","adapter":"openai"},"maker":{"slug":"openai","display_name":"OpenAI","country":"US"},"country":"United States","compliance":{"residency":"US","country":"United States","inference_location":"US","zdr":{"logging":true,"moderation":true,"caching":true,"training":false,"subprocessors":null},"training":"no","logging":"abuse_monitoring","retention_days":30,"moderation":"default_on","caching":"implicit","cache_scope":"organization","cache_ttl_seconds":1800,"content_storage":"retained","dpa_available":true,"third_party_access":true,"transfer_mechanism":"sccs","route_id":"openai/api-us"},"release":{"date":"2025-10-06","source":"vendor","url":"https://developers.openai.com/api/docs/changelog"},"references":[{"kind":"model_card","url":"https://developers.openai.com/api/docs/models/gpt-realtime-mini","created_at":"2026-10-07T05:57:44.064854Z","updated_at":"2026-10-07T05:57:44.064854Z"}],"input_modalities":["text","image","audio"],"output_modalities":["text","audio"]},{"id":"openai/gpt-realtime-translate","type":"realtime","name":"GPT Realtime Translate","model_id":"gpt-realtime-translate","capabilities":["audio","text"],"speed":"fast","quality":"best","context_window":16000,"max_output_tokens":2000,"description":"OpenAI's real-time speech-to-speech translation model over a dedicated realtime endpoint — auto-detects the source language and speaks a configurable target language while you keep talking. Set translation_target_language (BCP-47) on session.start. Continuous streaming, billed by audio duration.","params":{"realtime":{"input_audio_format":"pcm16","output_audio_format":"pcm16","sample_rate":24000,"input_sample_rate":24000,"output_sample_rate":24000}},"pricing":{"billing_unit":"per_minute","rates":[{"unit":"per_minute","amount":0.034,"dimensions":{"direction":"input"}}],"audio_input_per_minute":0.034},"featured":true,"region":"US","model":"gpt-realtime-translate","family":"gpt-realtime-translate","pooled":true,"provider":{"slug":"openai","display_name":"OpenAI","owner_country":"US"},"service_route":{"id":"openai/api-us","adapter":"openai"},"maker":{"slug":"openai","display_name":"OpenAI","country":"US"},"country":"United States","compliance":{"residency":"US","country":"United States","inference_location":"US","zdr":{"logging":true,"moderation":true,"caching":true,"training":false,"subprocessors":null},"training":"no","logging":"abuse_monitoring","retention_days":30,"moderation":"default_on","caching":"implicit","cache_scope":"organization","cache_ttl_seconds":1800,"content_storage":"retained","dpa_available":true,"third_party_access":true,"transfer_mechanism":"sccs","route_id":"openai/api-us"},"release":{"date":"2026-05-07","source":"vendor","url":"https://developers.openai.com/api/docs/changelog"},"references":[{"kind":"model_card","url":"https://developers.openai.com/api/docs/models/gpt-realtime-translate","created_at":"2026-10-07T05:57:44.064854Z","updated_at":"2026-10-07T05:57:44.064854Z"}],"input_modalities":["audio"],"output_modalities":["text","audio"]}],"total":5,"offset":0,"limit":50}