Multimodal

Example	Description
Audio Input Output	Send a WAV recording to gpt-audio and save the model’s spoken reply to tmp/result.wav with write_audio_to_file().
Audio Sentiment Analysis	Analyze speaker sentiment in a recorded conversation with Gemini, then ask a follow-up question answered from SQLite-backed session history.
Audio Streaming	Stream pcm16 audio from gpt-audio and write the chunks to a WAV file while printing the transcript.
Audio to Text	Transcribe an MP3 conversation with Gemini, labeling each speaker in the output.
Image To Audio	Convert image descriptions to audio output.
Image To Image	Transform images using agent-driven processing.
Image To Structured Output	Extract structured data from images.
Image To Text	Pass a local sample.jpg to GPT-5.2 with OpenAIResponses and print a three-sentence story about the image.
Media Input For Tool	Example showing how tools can access media (images, videos, audio, files) passed to the agent.
Video Caption	Generate captions from video content.