Cloud providers

Cloud LLM providers are optional during soft launch. Ollama alone is enough for a free solo workflow. Add cloud keys when you want stronger models, faster replies, or better tool-calling for complex Simulink edits.

Supported providers

MBD Mind supports multiple cloud backends, including:

  • OpenAI
  • Anthropic
  • Google Gemini
  • Groq
  • DeepSeek
  • Kimi (Moonshot)
  • Mistral
  • Cohere

Availability may vary by build; open Settings → AI Providers to see the full list.

Add an API key

  1. Obtain an API key from your provider's console.
  2. In the desktop app, open Settings → AI Providers.
  3. Expand the provider card.
  4. Paste the API key (and base URL if you use a proxy or Azure-style endpoint).
  5. Click Save, then Verify or Refresh models.
  6. In the model toolbar, select the provider and model.

Keys are stored locally on your machine, not in chat logs sent to MBD Mind servers.

When to use cloud vs Ollama

Use OllamaConsider cloud
Everyday Q&A and explorationLarge model_edit sequences
No API spendWeak local tool-calling
Strict data localityTime-sensitive deadline work

You can switch per message from the model toolbar without restarting the app.

Budget and local-only guardrails

In AI Providers settings:

  • Set a monthly budget and warning threshold for cloud spend
  • Enable local only to block all cloud LLM calls (Ollama still works)

Usage totals appear in the Usage dashboard and sync to your org for admin reporting on team plans.

Security notes

  • Do not commit API keys to git or share them in agent prompts.
  • Rotate keys if a machine is lost or imaged.
  • For IT-managed teams, org admins may publish a default Ollama URL instead of distributing cloud keys.

Troubleshooting

StatusAction
auth failedRegenerate key; check for trailing spaces
quota exceededBilling or rate limits on provider side
rate limitedWait or switch provider/model
local only blockedDisable local-only mode or use Ollama

Next steps

Cloud providers | MBD Mind