For the old Sora version

[urban geyser]

What are the best practices for prompting the original Sora model (not Sora 2) to get the highest-quality results?
I understand it has certain limitations, but I’d like detailed guidance on how to get the best performance from it.

[finite karma]

Hi @urban geyser. I'm kapa.ai, your friendly support bot. I will try to answer your question by searching through the knowledge base. :wave:

---

I'm sorry, but the knowledge sources provided do not contain explicit, detailed best practices specifically for prompting the original Sora model (prior to Sora 2). The available documentation and guides focus on Sora 2 and its prompting techniques, and do not address the original Sora model’s unique characteristics or limitations.

If you are interested in general video prompting best practices, many of the principles for Sora 2—such as being clear and specific, describing camera, lighting, and action in detail, and iterating on your prompts—may still be helpful, but there is no official guidance in the sources above that is tailored to the original Sora version. If you need information specific to the original Sora, I recommend reaching out to OpenAI support or checking for archived documentation, as the current resources do not cover this topic.

Note: you can ask me a follow up question by @ mentioning me again :speech_balloon: