Hi everyone,
I’ve been experimenting with the new web search feature (currently in preview), and I’m pretty concerned about how quickly the costs add up. My use case involves larger text generation that needs to be re-checked multiple times in a single prompt. Because of how often the browsing feature is invoked, the estimated cost has shot up far beyond what I expected.
Do you think/know if OpenAI plans to reduce costs once the feature moves out of preview, or is this what we should expect long-term?
I’m at a crossroads: if pricing stays this high, I might need to pivot to a different pipeline. If any of you have tested alternatives and found a workable hybrid, I’d love to hear about it.