A fresh project hit Hacker News yesterday with a straightforward pitch: what if you could query GPT, Claude, and Gemini simultaneously without managing three separate subscriptions? Whizi.io posted their "Show HN" launch around 5:20 AM UTC on August 15th, drawing modest attention with a score of just 4 points and zero comments at publish time. The core concept is aggregation—give users one interface to ask questions across all three major frontier models while paying for access through a single subscription tier.

What's Actually Being Sold Here

The premise is straightforward enough: Whizi positions itself as a unified query layer sitting atop OpenAI's GPT family, Anthropic's Claude lineup, and Google's Gemini series. Instead of maintaining separate API keys and billing relationships with each provider, users would theoretically route their prompts through Whizi and receive responses from whichever models they want to compare or combine. The tagline "Ask them all. Pay for one" suggests a pricing model where the service bundles access in a way that could save money versus going direct—but concrete rates aren't visible in the HN post itself.

Why This Category Keeps Popping Up

Multi-model aggregation isn't exactly novel territory. Various startups and side projects have attempted to build aggregator layers over the past two years, typically targeting power users who want to compare outputs or leverage different model strengths without context-switching between ChatGPT, Claude.ai, and Gemini interfaces. The challenge has always been margin compression: when you're just wrapping APIs from three well-capitalized AI labs, there's not much room for sustainable pricing unless you add significant value through workflow integration, caching, or specialized fine-tuning. Whizi will need to articulate what differentiates their wrapper beyond convenience.

What We Don't Know Yet

The HN post gives us the concept and the URL but precious little technical detail. No mention of which specific model versions are supported—whether we're talking GPT-4o versus older variants, Claude 3.5 Sonnet versus Opus, or Gemini 1.5 Pro. No pricing page is visible from external crawling, so potential users can't evaluate whether "pay for one" actually translates to competitive costs when you factor in Whizi's own markup. And critically, zero comments on the HN thread suggests either the post flew under the radar or early visitors found nothing worth discussing yet.

The Technical Angle Worth Watching

If Whizi is doing real-time fan-out queries—sending your prompt to all three providers simultaneously and returning aggregated results—that's architecturally interesting but operationally complex. Latency becomes a significant factor when one model might respond in two seconds while another takes eight. There's also the question of how they handle streaming responses, context window management across different providers, and whether they're doing any post-processing on outputs before returning them to users.

Key Takeaways

  • Whizi.io launched August 15th as an HN Show project with a simple aggregation pitch
  • The service claims to let users query GPT, Claude, and Gemini through one interface and subscription
  • Specific model versions, pricing tiers, and technical implementation remain unconfirmed from public sources
  • Early community reception appears minimal based on HN scoring and comment count

The Bottom Line

Whizi's pitch is clean, but the aggregator market has a brutal track record—convenience alone rarely justifies a markup when the underlying APIs are already accessible. Until they publish pricing and technical specifics, this feels less like a disruptive platform and more like an interesting experiment that needs more runway to prove its case.