English

Deutsch

English

Deutsch

Appearance

Models ​

nunq supports multiple AI model providers simultaneously, giving you flexibility to choose the right model for each task. Administrators configure which providers and models are available; users can then select from the enabled models when chatting.

Supported Providers ​

nunq integrates with the following AI model providers:

  • Azure OpenAI — Microsoft-hosted OpenAI models with enterprise compliance and data residency options.
  • OpenAI — Direct access to OpenAI models.
  • Anthropic — Claude models from Anthropic.
  • AWS Bedrock — AI models hosted on Amazon Web Services.
  • Google Vertex AI — Google's enterprise AI platform, including Gemini models.
  • Ollama — Self-hosted open-source models for on-premises deployments.
  • Mistral — Models from Mistral AI.

Model Capabilities ​

Models in nunq are categorized by their capabilities:

Chat Models ​

Standard conversational models used for general-purpose text generation, Q&A, summarization, translation, and other language tasks. These are the models you interact with in the chat interface.

Reasoning Models ​

Advanced models that perform explicit chain-of-thought reasoning before producing an answer. These models are well-suited for complex analytical, mathematical, and logical tasks. When using a reasoning model, you can view the reasoning steps in the chat. See Chat Interface for details on the reasoning display.

Image Generation Models ​

Models that create images from text descriptions. When image generation is enabled, the AI can produce visuals based on your prompts.

Embedding Models ​

Models used internally for knowledge source indexing. Embedding models convert text into numerical representations (vectors) that enable semantic search across your knowledge sources. These models are not used directly in chat but power the retrieval-augmented generation (RAG) pipeline. See Knowledge Sources for more information.

Speech-to-Text Models ​

Models that transcribe spoken audio into text. These are used by the voice recording feature in the chat interface to convert your speech into a text message.

Browsing Available Models ​

Navigate to the Models page to see all models available in your organization. Each model listing includes the provider, model name, and supported capabilities.

Selecting a Default Model ​

You can set a default model for your account from the Models page or your account settings. The default model is pre-selected whenever you start a new chat, saving you from choosing a model each time.

Usage Tracking ​

nunq tracks your token consumption per model. You can view your usage statistics to understand how many tokens you have consumed across different models and time periods. This is useful for monitoring costs and understanding your usage patterns.

Usage Limits ​

Administrators can set usage quotas to control token consumption. Limits can be configured on an hourly or daily basis per model. When you reach a limit, you will be notified and will need to wait for the quota to reset or switch to a different model. This helps organizations manage costs and ensure fair resource allocation across users.