Choose your models, keep your freedom
You ask: How do we avoid depending on a few commercial AI providers?
Models keep getting better, and you do not want to be tied to one supplier.
You choose which AI models do which work, and where they run: commercial, open, European or your own machine. Changing a model is a setting, not a rebuild, so you can switch as better ones appear. Your code and records stay in your own code host. Today every decided role runs on one provider.
The freedom to choose is built. It is not yet used, and no European route is connected.
How it answers
- The model for each role is a setting behind one door, not code. Measured
- Three providers answer, two are wired to the gateway, and all four decided roles run on Anthropic today. Built
- An open model moves into a role only after it is tested against a top commercial model on the same test set. Planned, decided 2026-10-13
- The European route, in order: reviewers on Scaleway, then all models, then hosting. The Scaleway key answers (16 models) and is not connected yet. Planned, decided 2026-10-13
- Your code host stays the single record of your work, and B Team shows you a view of it. Planned, not built
- You can run it on a MacBook with Docker Compose. A guided setup for your own account or server is planned. Built (MacBook); server not
Reliability on this page
This is the "tested" leg: a model or an instruction change is used only after it passes its test set, and open models are measured against a top commercial one (claims 20 and 60). How we raise reliability
The honest limit
- The reviewer is not yet a different model family from the worker (issue #82).
- The Scaleway key was exposed in a transcript on 2026-09-24 and is replaced before it is connected (issue #81).
- Today you need GitLab. Other code hosts and self-managed GitLab are proposed (issue #279).
- Other tools are ahead on this: Lovable, Linear, Copilot and Productboard (in beta) already offer hosting in EU regions. Ours is later. Where other tools are ahead is on the page of what is not built.
- A hosted service is an intention only: we intend to offer a hosted service in Europe. It needs secrets kept out of git, prompts and logs, a full audit log, a data policy checked against each provider's terms, separation per organisation, and sign-in. None is measured today.
- "Hosted in Europe" is not the same as "data stays in Europe". The model providers' terms decide where prompts go.
For NGOs
Digital sovereignty, no lock-in, and the freedom to change supplier.
The technical detail
The claims behind this page, with their states as the evidence page records them.
| No. | Claim | State | What it serves | Check or issue |
|---|---|---|---|---|
| 2 | Every model call goes through one door. A role's model is a setting. | Measured | One gateway (F2) | gateway_checks |
| 3 | Not tied to one model vendor. Three providers answer, two are wired to the gateway, and every decided role runs on one. | Built | Independent of any vendor (vision) | Three providers answer, two wired, one used |
| 12 | The reviewer is a different model family from the worker. | Not built | Different family by default (F13) | #82 |
| 14 | One inbox for everything that needs a person. | Built (per project) | One review inbox (F8) | across projects #111, #31 |
| 20 | A test set must pass before a model or instruction change is used. The gate is a person's command. | Measured; gate manual | Test-set gate (F5) | golden_checks; #115 |
| 44 | Runs on a laptop or a small server. | Built (MacBook); server not | Host profiles (N1) | #44 |
| 50 | The code host is the single record, behind one layer. Self-managed GitLab first. | Not built | Code host as record (F25) | #279; #289 |
| 53 | A guided setup into your own account or server. | Not built | Guided setup (F28) | #291 |
| 54 | You may pick only models your data policy allows that passed the test set. | Not built | Model choice bounded (S21) | #294 |
| 55 | A hosted service in Europe separates each organisation. Intended only. | Not built | Separation per organisation (S22) | #295 |
| 59 | Models run in Europe in steps: reviewers on Scaleway, then all models, then hosting. The Scaleway key answered 200 and listed 16 open models on 2026-10-04. It is not connected to the engine, and it is replaced (#81) before it is. | Not built | Model choice bounded (S21), hosted later (S22) | Decided 2026-10-13. Not built. #294 |
| 60 | A role moves to an open model only after a test-set run against a top commercial model. The nearest built part is the test-set gate (claim 20). | Not built | Test sets (F5), open against commercial | Decided 2026-10-13. Not built. #314 |
The other four questions
-
Question 1
Build what actually matters
Start from your goal. End with real users. Judge by use.
Read the answer -
Question 3
People and AI, building together
Agents write. People decide, guide and check.
Read the answer -
Question 4
Use AI and stay compliant
B Team produces the records your compliance needs as you work.
Read the answer -
Question 5
Fit your people and your budget
Work follows your rhythm. Much of it runs in batches.
Read the answer
Tell us you are interested
B Team is not offered to anyone yet. NGOs come first. If you want to hear when that changes, or you would like to be one of our first pilots, write to us.
Write to: [contact address to be decided]
This draft has no form and collects no data. Once there is an address, what happens next is meant to be this: we tell you whether we can work with your organisation yet. We promise no date and no price.