Skip to content
ModelOpt

Getting Started

Follow these steps to run your first optimization report.

  1. 1. Open the Optimizer page and select your GPU.
  2. 2. Set system RAM and confirm VRAM override.
  3. 3. Pick at least one use case and speed preference.
  4. 4. Submit and review top model recommendations.

Hardware requirements explained

Local deployment mode enforces VRAM and RAM thresholds. Cloud mode prioritizes RAM constraints and use-case fit.

Troubleshooting quick tips

  • - Ensure GEMINI_API_KEY is configured.
  • - Retry after one minute if rate limited.
  • - Use cloud mode if local VRAM is below model minimum.

Next steps