--- title: "Cloud AI Providers · AssistantGeneral" description: "Bring your own key for Claude, OpenAI, Gemini, Grok, Groq, Mistral, DeepSeek, Together, OpenRouter or any OpenAI-compatible server. Opt-in, never the default." url: "https://royalsoftworks.com/products/assistant-general/features/cloud-ai-providers/" source: "https://royalsoftworks.com" format: "markdown" note: "Markdown rendering of the HTML page at `url`. Same content, same canonical URL." --- Features · The AI itself # Use a cloud model only when you choose to. Bring your own key for Claude, OpenAI, Gemini, Grok, Groq, Mistral, DeepSeek, Together, OpenRouter or any OpenAI-compatible server. Opt-in, never the default. [Editions](https://royalsoftworks.com/products/assistant-general/editions/): Every edition; Free uses its own key. Custom headers and a model server elsewhere in your building: Pro and up. ## Cloud AI providers, feature by feature Each comes with a small drawing of how it works. They are diagrams, not screenshots; numbers in them are examples. - ### Nine providers by name Anthropic (Claude), OpenAI, Google Gemini, xAI (Grok), Groq, Mistral, DeepSeek, Together AI and OpenRouter, picked from a list. Diagram: Claude, OpenAI, Gemini, Grok, Groq, Mistral, DeepSeek, Together, OpenRouter. - ### Any OpenAI-compatible server Your company's own model server, a gateway, or a self-hosted endpoint: a base address, a model name and a key. Diagram: Address → Model name → Key → Connected. - ### Your key, your bill You pay the provider directly. Nothing is routed through us, and the local model stays the default. Diagram: Your key → Provider → Your invoice. - ### A model server down the hall Point it at a model running on another machine in the building, with no sign-in, a key, or a user name and password. Diagram: This desk → Local network → Model server (key or login). - ### Thinking effort For models that reason before answering: the provider's default, or low, medium or high effort. Scale from Low to High, set at “Medium”. - ### The context window, read or set It reads how much text the model can hold from the service, and you can override it by hand. Diagram: Reported by service and Set by hand come together as Context window. - ### Custom HTTP headers Headers sent with every chat request, for gateways and proxies that need them. Diagram: Chat request → + your headers → Gateway → Model. - ### The administrator decides On a managed desk, the policy says whether cloud models may be used at all, or whether everyone goes through the company server's model. Diagram: Cloud request, then “Policy”: yes, Allowed; no, Local only. ## Goes well with - [Local AI models](https://royalsoftworks.com/products/assistant-general/features/local-ai-models/) - [Model cluster](https://royalsoftworks.com/products/assistant-general/features/model-cluster/) - [Policy controls](https://royalsoftworks.com/products/assistant-general/features/policy-controls/) [Previous Local AI models](https://royalsoftworks.com/products/assistant-general/features/local-ai-models/) [Next Model cluster](https://royalsoftworks.com/products/assistant-general/features/model-cluster/) [All 31 feature classes](https://royalsoftworks.com/products/assistant-general/features/)