Hi HN I’m Tina. Like many of you, I’ve been frustrated by LLM hallucinations — outputs that sound perfectly confident but are fabricated. So I built CompareGPT.io, a tool that: Runs the same query across multiple LLMs (ChatGPT, Gemini, Claude, Grok, etc.) Shows results side by side so discrepancies (and hallucinations) are easier to catch Offers a unified API for easier integration We just opened a waitlist (with a half-price early offer + lifetime plan giveaway). I’d love feedback from this community: do you think multi-model consistency could become a standard way to reduce hallucinations in law, finance, or research? Thanks!