Local LLMs
Execute LLMs directly within a DaDesktop instance, eliminating the need to route prompts through external APIs. By leveraging DaDesktop's GPU resources, you retain full control over your data, ensuring that prompts, files, and other assets remain confined to your private environment. This model offers a cost-effective alternative, as you are billed for the desktop instance rather than on a per-token basis.
Benefits of Local Execution
- Enhanced Privacy: All data, including prompts and files, remains securely within the DaDesktop environment.
- Predictable Costs: Utilize DaDesktop's GPU infrastructure without incurring additional charges for each API request or token processed.
- Model Autonomy: Select the specific LLM you wish to deploy and fully manage its configuration.
- Integrated Workflows: Execute models directly from the desktop, seamlessly alongside other applications and tools for custom workloads.
AI Agents
Deploy agents powered by local models to automate tasks and interact with various tools.
Coding
Leverage local models to assist with software development, coding, and testing processes.
Research and Experimentation
Conduct in-depth research and analysis, including the evaluation of different models and configuration settings.
Mechanism
To get started, select a DaDesktop GPU-enabled instance, load your preferred model, and initiate local execution. The GPU passthrough feature ensures exclusive access to a physical GPU, optimized for LLM workloads. For detailed specifications and supported configurations, refer to the GPU page.
Start Your Free Trial Today
Run seamless virtual IT training with cloud-based labs, no downtime, just scalable learning that works.