> ## Documentation Index
> Fetch the complete documentation index at: https://vokra.ai/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Model selection

> Add provider credentials and choose a model per agent.

Each agent uses one workspace LLM credential and one model from that provider. Add credentials in **Settings → LLM credentials**, then select the credential and model while creating or editing the agent.

The current provider registry includes OpenAI, Anthropic, Google AI, Groq, Mistral AI, AWS Bedrock, Azure OpenAI, Ollama, DeepSeek, Hugging Face, xAI, and OpenRouter. The app displays the current fields and model options for each provider.

You supply the provider account and credential. Provider charges, availability, data handling, rate limits, and model retirement policies are separate from Vokra. Test model changes on a manual run before relying on scheduled jobs.