A Private AI Workbench for Your Team
Your team's AI, in one workbench, deployed on your own server — company data stays in-house, and accounts, permissions and usage are managed centrally. Delivered on the open-source MonkeyCode (Baizhi) workbench.
What it is
MonkeyCode (Baizhi Workbench) is an open-source AI workbench for teams: members describe a task in plain language and the AI completes office, coding and content work and delivers the result. The team shares models, a knowledge base, skills and tools, while an admin console manages member permissions, usage statistics and resources. We deploy it on your server, complete the initial setup, and provide ongoing upgrades and maintenance. The software is licensed under AGPL-3.0.
Who it suits
- Companies that want their team using AI, without putting company material into public platforms
- Teams that already have a cloud or office server and want their data to stay on it
- Teams of up to 10 people (covered by the free tier), or larger teams willing to buy the vendor's licence for more seats
What we do
- Deployment: installation and setup on your cloud or office server
- Initial configuration: admin account, member seats and permission groups set up once
- Model connection: connected to the model service you choose (model API fees are paid by you directly to the model provider)
- Knowledge base and tools: initial configuration based on your business materials
- Handover training: one remote session each for admins and members, with a short usage guide
- Ongoing maintenance (yearly): version upgrades, backup checks and day-to-day Q&A
How pricing works
- The software: the vendor's free tier covers up to 10 seats at no software cost; beyond 10 seats a licence must be purchased from the vendor, and we can help arrange it
- Our service fee: deployment and initial configuration from ¥1,980 (one-off); yearly maintenance from ¥2,980/year
- Server and model API costs are yours; we advise on sizing for your team (the vendor's light configuration starts at 1 CPU core + 1 GB RAM)
Where your data lives
The workbench and your team data are stored on your own server, with access and backups under your control. When an external model is used, the content you send to the model goes to the model service you configured — which model and which permissions are your decision, and we configure them as you specify.
Try the demo first
The vendor runs a public demo you can try yourself: open the demo workbench (email code sign-in), or the demo admin console (account admin@baizhi.cloud, password work.baizhi.cloud — the vendor's published demo account). It is a public environment, so please do not enter real company material into the demo.
Free enquiry →