Software
You can run the model yourself
Weights and tools are public. You are not renting every token from one cloud vendor forever — if you own the stack.
Open Source AI · Paid help · Canada
Open source AI is free. Getting it right is not.
Models like Llama, Mistral, and local runtimes can run on your hardware with less vendor lock-in. Most teams still need install, security, RAG, and staff training — that is the paid work we do from Alberta.
Pricing honesty: open-source software licenses are often free. Our paid packages cover setup, hardening, document search (RAG), support, and training. Quotes after a free 30-min call.
What “open source AI” means for a business
Control
Data can stay in Canada / on-prem
Ideal when client files, industrial data, or privacy rules make public ChatGPT the wrong default.
Reality
Free ≠ free-to-operate
Hardware, updates, access control, backups, and staff habits cost time. We sell that expertise.
What we help with (paid)
Setup
Local / private stack
Ollama or similar runtimes, model choice, GPU/CPU sizing, secure access for staff, basic document Q&A on your files.
Hybrid
Open source + cloud tools
When Claude/Copilot is fine for some work and local models for sensitive work — with clear rules so staff do not mix them up.
Training
Teach the team
How to prompt, when to stay local, what never to paste into public AI — private coaching or team workshops.
Support
Keep it alive
Updates, new models, access for new hires, and a human when something breaks — optional retainers after setup.
Packages (CAD · indicative)
| Package | Best for | What you get | From |
|---|---|---|---|
| Discovery call | Anyone exploring open source | 30 min map: tools, risk, first step | Free |
| Open-source pilot | One team, one workflow | Local model install, secure access, 1 workflow, staff demo | Quoted |
| Private AI deploy | Sensitive data / on-prem | Hardened local stack + document search path | Quoted |
| Training add-on | After seats or local install | Private coaching or half-day workshop | Academy rates |
No fake “$X forever” claims — scope depends on hardware, users, and data sensitivity. Fixed quote after the free call.
Open source vs paid cloud (simple)
| Question | Open source / local | ChatGPT / Claude / Copilot |
|---|---|---|
| Monthly seat cost | Often $0 license; hardware & support cost | Recurring seats / usage |
| Data leaves building? | Can stay local if designed that way | Usually vendor cloud (check contracts) |
| Setup skill needed | High without help | Low for chat; higher for agents |
| Best first move | Pilot or pilot with us | Training + usage rules |
Want open source AI without the weekend of YouTube?
Book a free discovery call. If a local or hybrid stack fits, we quote a fixed pilot. If cloud tools are enough, we say so and train your team instead.