No recurring bill, by design
The AI runs in the visitor's browser. We don't pay per-inference, so we don't bill you per-message — it's structural, not a promotion.
Every other AI chat widget bills you forever. Zupport.chat doesn't. There's no subscription, no usage meter, no OpenAI key to provision. You pay $79 once, paste a snippet, and the widget keeps working with no further bill.
The AI runs in the visitor's browser. We don't pay per-inference, so we don't bill you per-message — it's structural, not a promotion.
Most “AI chatbot” tools make you bring your own OpenAI key. That's how the bill sneaks back in. Zupport.chat has no external API call.
Subscription tools cap your conversation count and bill overage. Zupport.chat has no meter to overage on.
Sourced from each company's published pricing page. Your numbers will vary; the ratio won't.
| Zupport.chat | Subscription AI widgets | |
|---|---|---|
| Monthly fee | $0 | $19–$299/mo |
| Per-message fee | $0 | Common (or counted against quota) |
| OpenAI API key needed | No | Often yes |
| Conversation quota | None | Plan-capped |
| Cost in year 3 | $79 total | $684–$10,764 total |
The model runs on the visitor's device via WebGPU. We don't pay per-message inference, so we don't need a meter. Our only recurring cost is the small VPS that stores your knowledge base config — covered by the one-time fee's margin.
For ecommerce support, sizing, returns, FAQ — no, it's comparable. The model (Qwen3) is small but fine-tuned for retrieval-augmented Q&A. For long-form reasoning across many documents, hosted models still win.
Mobile and slow devices get the scenario layer (instant deterministic answers for the questions you authored). The AI tier degrades gracefully to retrieval-summary on weak hardware.
No. The $79 covers everything Zupport.chat-side. You might pay your hosting provider, but that's your hosting, not us.
Founder pricing for the first 100 customers, then $99 for everyone after. Both are one-time.
One payment. AI included. No monthly fee. Forever.