Hey PH 👋 I'm Muhammad, the solo developer behind Versus AI. The idea came from my own frustration — I was constantly switching between Claude.ai, ChatGPT, and Grok to compare responses to the same question. One day I just decided to fix that. The hardest technical part was getting three different streaming APIs to fan out simultaneously through a single SSE connection. Each provider has slightly different streaming formats, different auth, different quirks — Grok uses max_tokens, OpenAI uses max_completion_tokens, they all have different role naming conventions. Getting them all to feel seamless was about 80% of the actual work. A few things I'm genuinely proud of: Responses stream in real time as the model generates them, not buffered Markdown renders correctly during streaming, not just when done Image comparison is side by side with retry per model if one fails The whole thing is native SwiftUI — no Electron, no web views Would love any feedback, especially on the model selection UX — that's the part I'm least sure about. Try it free, search Versus AI on the Mac App Store

Versus AI

Compare Claude, GPT, and Grok side by side in real time

Compare Claude, GPT, and Grok side by side in real time