How it Works — Real Use Cases
No account. No separate install. No Big Tech dependency. Every General Bots app ships pre-installed in the suite and runs on infrastructure you own — this is how it works and what teams actually use it for.
Deploy the whole suite on your VPS
One container, every app pre-installed and pre-integrated. Your data stays on your hardware — no accounts, no SaaS gatekeeper.
Open it from the desktop launcher
The app is one click away inside the General Bots desktop. No separate signup, no setup wizard, no credit card.
Drive it with plain language or the API
Ask the AI assistant to do the work — or call the REST API directly from your own tools. The endpoints below are real, from the source code.
Try these real endpoints
/api/vision/analyze/api/vision/describe-video/api/vision/history/api/vision/qrcodevision.teach.case1_t
vision.teach.case1_d
vision.teach.case2_t
vision.teach.case2_d
vision.teach.case3_t
vision.teach.case3_d
Built on the Open-Source Code
Real technical data extracted from the Rust source repository — API endpoints, database tables, and code size. Every number on this page is auditable in the code.
Real API surface (from the source)
/api/vision/analyze/api/vision/describe-video/api/vision/history/api/vision/qrcodeDatabase schema (from migrations)
Everything a Modern Vision Needs
Powerful capabilities with AI orchestration — all self-hosted on your infrastructure.
OCR Text Extraction
Extract text from scanned documents and handwriting in multiple languages, with structured output and per-extraction confidence scores.
Object Detection
Detect and count objects in images with bounding-box coordinates and confidence scores for every identification.
Damage Analysis
AI-powered damage assessment with severity classification, repair cost estimation, and generated assessment reports.
License Plates
Read plate numbers, cross-reference vehicle databases, support multi-country plate formats, and track recognized plates over time.
Analysis History
Browse, search, export, and share every past analysis from the history panel — each result retained with full tracking.
Model Flexibility
Choose the right AI model per capability — OCR, objects, damage, plates — via the vision models endpoints.
One Flat Server — Zero Per-User Fees
Unlike commercial SaaS suites that charge per seat, General Bots runs on your own infrastructure with unlimited users.
- Unlimited users and operations
- No per-user licensing fees
- Full source code access (AGPLv3)
- Self-contained Docker deployment
- Bring your own LLM / AI backend
Plus overage fees
On your infrastructure
Self-Hosted vs SaaS
See how the open-source General Bots Vision compares to proprietary alternatives.
| Feature | Self-Hosted | SaaS |
|---|---|---|
| Per-user pricing | Free | Vendor lock-in |
| Data ownership | Yours | Theirs |
| Source access | Full (AGPLv3) | Closed |
| AI model choice | Any LLM | Vendor lock-in |
| Scalability | Unlimited | Tiered |
| Audit & compliance | Full control | Limited |
Built for Sovereignty, Scale, and AI
Thousands of teams choose General Bots for their vision infrastructure.
Zero Vendor Lock-In
Your data, your infrastructure, your control. General Bots is fully open source under AGPLv3 — no proprietary APIs, no data leaving your network, no forced upgrades.
- 100% Open Source
Audit, modify, and extend the code. Community contributions welcome.
- AI-Native by Default
LLM orchestration built in — bring any model, run it on your own hardware.
Related Reading
The Illusion of Intelligence
Statistics vs. understanding — a didactic deconstruction of modern LLM architectures.
Read MoreDigital Twins: Architectural Replicas of Business Logic
How digital twins — powered by sovereign AI orchestration — are transforming personal productivity and industrial system monitoring.
Read MoreWhat Is an LLM? A Practical Guide
Tokens, parameters, RAG, and the sovereign AI pivot—everything you need to understand Large Language Models
Read MoreReady to Deploy?
Get your own self-hosted vision platform running in minutes. No per-user fees, no vendor lock-in.