← Back to landing

Setup guide

Local vs Cloud vs Hybrid

Pick your install strategy based on the computer you actually have — not the one you wish you had.

ClickSorcery runs Jerry through OpenClaw. You choose where the AI brain lives: on your hardware (Ollama), in vendor clouds (direct APIs), or through a smart router (OpenRouter / DeepInfra). Telegram and OpenClaw software stay free — you only pay when thinking happens in someone else's cloud.

Three architectures

TraitStrategy A — Local (Ollama)Strategy B — Direct cloud APIsStrategy C — Hybrid (OpenRouter / DeepInfra)
Computer neededStrong workstation — VRAM mattersNormal laptop (8–16 GB RAM is enough)Normal laptop — compute lives in the cloud
Fixed monthly cost€0 API (electricity only)High & unpredictable — each vendor bills separatelyVery low — pay per token you actually use
Data privacy100% on your machineEnterprise contracts possible, but data crosses to cloud regionsMedium — traffic passes brokers; not for raw banking data
Biggest upsideNo internet dependency for inferenceOriginal Claude / OpenAI environmentsBest price flexibility on the market
When to choose16 GB+ RAM or dedicated GPU, privacy-first8 GB laptop, need top-tier single vendorDefault sweet spot for most users
Strategy A

Local computer + local models

You have a real AI workstation and want zero API bills and full privacy.

ModelHardwareGoodAvoid when
Qwen3.5:4b or 9b16 GB RAM laptop or 6 GB+ VRAM GPUVery fast, low memory — good for routine sub-agent tasksStruggles with very hard logic and long code debugging
Gemma4:26b-moe (or similar MoE)16–24 GB VRAM (RTX 3090/4080) or Mac 36 GB+ RAMMoE architecture — smart and fast for its sizeHeavy RAM/VRAM — can freeze a normal office PC
DeepSeek-R1:32b or 70bAI workstation — dual RTX 4090 or Mac Studio 64–128 GBWorld-class reasoning and code — €0 per message at homeHuge model files (40 GB+), slow load without top hardware
Strategy B

Normal PC + direct APIs (Claude, OpenAI)

You need corporate-grade vendors, official channels, and absolute top models — on any laptop.

ModelHardwareGoodAvoid when
Claude Sonnet (Anthropic API)Any PC — even an old 8 GB laptopExcellent coding and long-context work in CursorExpensive — long chats burn balance fast (real users: ~€5 / ~10 questions + research)
OpenAI o1 / o3-miniAny PCStrong math and step-by-step reasoningClosed stack, data to OpenAI, separate billing
GPT-4o-miniAny PCFast, relatively cheap API baselineLess capable than Sonnet or o1 for hard tasks
Strategy C — recommended

Normal PC + OpenRouter / DeepInfra

Sweet spot: one API key, many models, penny-scale daily use.

ModelHardwareGoodAvoid when
DeepSeek-R1 (671B via router)Any PC — cloud runs the modelTop-tier reasoning for cents — great demo of agent powerSlow replies while it “thinks”
DeepSeek-V3 or Llama-3.1-70BAny PCBest price/speed — chat all day for a few centsNot offline; data leaves your machine
Claude Sonnet (via OpenRouter)Any PCCall Sonnet through same wallet as cheap modelsBackground agents can drain balance if unchecked

What to pick in Magic Installer

Under 16 GB RAM, no GPUCloud CommanderDo not choose Local Supercomputer — Ollama often never answers
16 GB RAM, no GPULocal possible (slow) or Cloud Commanderqwen2.5:7b class — OK for light Jerry + Telegram
8–16 GB VRAM GPULocal SupercomputerJerry + several agents, €0 API
24 GB+ VRAMLocal Supercomputer (full company)Optional cloud API for hardest tasks only

Practical launch formula (hybrid + optional local)

  • Daily chat & Telegram: OpenRouter or DeepInfra (DeepSeek-V3, Llama-3.1-70B) — often a few euros per month.
  • Stuck in Cursor on architecture: route one session through Claude Sonnet via OpenRouter.
  • When you buy a strong PC later: run Ollama locally for routine agents; keep cloud for critical tasks only.