Slowest, Most Expensive GPT and Its Performance Issues

Viewed 71
The project is currently experiencing significant performance challenges due to API usage limits and high demand. As the creator scales up from 2 to 5 VPCs, they've faced operational hurdles, including exceeding quotas from Anthropic and Google. While aiming to reduce costs (10-30 cents per search), the service remains slow. User interactions highlight a mix of amusement and skepticism, with some queries yielding redundant results similar to popular LLMs. Suggestions for improvement include a clickable marquee of examples to minimize costs and better manage user experience. Despite the challenges, there is interest in the potential for an open-source version and further discussions around creative prompts and responses.
0 Answers