Zapbit

Zapbit turns WhatsApp Web into an autonomous AI assistant powered by local LLMs running entirely on your own hardware — no cloud servers, no API keys, no data ever leaving your machine.

// HOW IT WORKS

01 · Configure your model
Install Ollama and choose your model: Gemma 4, Qwen 3, Llama 3, or any compatible open model. Zapbit connects to it locally — no subscriptions to AI providers required.

02 · Create a Skill
Crawl any website or upload your own documents to generate a custom SKILL.md. The bot learns your business rules, tone, products, and FAQs — then replies in character.

03 · Ready to go
Zapbit monitors WhatsApp Web, detects incoming messages, and replies automatically through a smart event-driven pipeline with serial FIFO queue and vision-based contact classification.

// KEY FEATURES

> LOCAL LLM — Ollama Integration
Run AI models directly on your hardware. No API keys, no per-message fees, no data leaving your machine. Your conversations stay yours — always.

> SMART AUTOMATION
Event-driven message detection keeps the bot responsive. A serial FIFO queue ensures conversations are handled one at a time, in order. Vision-based contact classification lets the bot read and identify chats intelligently.

> TEACHABLE SKILLS
The Skill Engine lets you train the bot with your own content. Crawl a website, upload PDFs or documents, and generate a SKILL.md that shapes how the bot thinks, responds, and represents your business.

// PLANS — One plan. Full power.

All plans include every feature. No tiers, no feature gates, no limits.

✓ Unlimited replies
✓ Local Ollama models (Gemma 4, Qwen 3, Llama 3 and more)
✓ File attachments
✓ Vision classification
✓ SKILL.md custom training
✓ Full privacy — 100% local processing

· Trial — Free (50 replies, no credit card required)
· Pro — $9.99/month
· Annual — $39.99/year ($3.33/month)

// PRIVACY

Zapbit is 100% local. Zero cloud. All processing happens on your machine. No message content is ever sent to external servers.