No fallback model available for gpt-audio and gpt-audio-mini #2305
Description
hi team i'm getting no fallback model available for the new gpt audio models via openrouter integration
i suspect we don't have integration to return audio yet, this is my code
<html>
<body>
<script src="https://js.puter.com/v2/"></script>
<script>
puter.print(`Loading...`);
puter.ai.chat(`What is 1+1?`, {
model: 'openai/gpt-audio',
}).then(puter.print);
</script>
</body>
</html>(openai claims it supports audio text input and output, but i tried to make it respond in text, still no fallback model)
i haven't tried directly calling openai for this, but i suspect this API is not a simple txt2speech or speech2txt, but more like an llm model that respons in audio (like a phone call?)
maybe our puter.ai.chat is evolving to allow more input and output, open to discussions