01 / Scheduled Tasks
Run recurring work on a schedule
Create scheduled tasks for morning briefings, weekly reports and follow-ups. Let local models handle recurring work in a consistent format.
View scheduled task details →Software Product
PINVOU AI Agent
Connect local models, files and tools in one workspace. Schedule tasks, reuse your team’s skills and view results and service status.
From a single task to a reusable workflow
Save recurring instructions, tools and schedules so you can reuse them for the next task.
- 01Describe the Task
Provide the goal, source material and output requirements
- 02Choose Models & Tools
Choose a model, tools and team skills
- 03Run & Review
Execute the task, check and refine the results
Bring local AI
into everyday work
Schedule tasks, connect business tools, reuse team skills and view service status. Start with a single task and gradually build workflows that fit your team.
02 / Tool Store
Bring business tools into your workflow
Install MCP integrations, API connectors, workflow tools and internal connectors in one place, so models can access information and carry out tasks within the permissions you grant.
View Tool Store details →03 / Custom Skills
Turn team methods into reusable capabilities
Save your team’s approaches to business writing, presentations, data analysis and visual design, so you do not have to repeat the context, role and format for every task.
View custom skill details →04 / Runtime Status
See how your local AI is running
View local computing resources, application services, model availability and throughput to understand resource usage, diagnose service issues and assess expansion needs.
View runtime status details →05 / Model Store
Model downloads
are now part of PINVOU
Model Store brings model management into PINVOU AI Agent. Choose models, download weights, deploy services and view runtime status in one workspace.
PINVOU / MODEL STORE
From validation reports to a working model
First check the model validation page for supported tasks, device counts, quantization formats and inference engines. Then manage models in PINVOU Model Store. The list shows models available to download, installed models and running models. After downloading, deploy and start the model from its card.
- Search by model, provider or inference engine.
- Filter by All, Installed or Running to check model status.
- Download, deploy and manage models from their cards.
Deployment & Connection
Connect to MegaCube model services
Use PINVOU on your everyday computer to access model services on MegaCube over the network, or run it directly on MegaCube if your deployment is configured for this.
- 01
Deploy a Model
Deploy the selected model and services on MegaCube, then check service endpoints and runtime status.
View validation records → - 02
Install PINVOU
Download and install the client that matches the operating system of the computer running PINVOU.
View downloads & connections → - 03
Configure & Verify
Configure the model connection for your actual network and permissions, then verify the result with a real task.
Open the user guide →