Skip to main content

​
What is Generative AI in UtopikAI?

UtopikAI is a model-agnostic platform. Generative AI powers chat, assistants, agents, workflows, Studio media generation, and knowledge search across your connected data and tools. You choose which LLM and embedding models fit each use case rather than being locked to a single provider. Supported chat providers include OpenAI, Anthropic, Google (Gemini), DeepSeek, Mistral, Meta (Llama), Alibaba (Qwen), xAI (Grok), and others.

See all LLMs

Browse the live catalog of language models available on UtopikAI, including capabilities and pricing.
Studio generation models are listed separately for image, video, and audio.

​
Where models are used

​
LLM vs embedding models

UtopikAI uses two related but distinct model types:
  • LLMs (large language models) — interpret user questions, read retrieved context, and produce responses. Used in chat, assistants, agents, and workflows.
  • Embedding models — convert documents and queries into vectors for semantic search. Used when indexing and querying connected knowledge sources.
For detailed provider lists and setup guidance, see LLM Providers and Embedding Models.

​
Why choose a different model?

  • Cost — Smaller or flash-tier models reduce usage cost for high-volume or simple queries.
  • Speed — Faster models improve responsiveness in chat and real-time workflows.
  • Task fit — Some models perform better for coding, reasoning, multimodal input, or domain-specific work.
  • Policy or deployment — Open-weight or region-specific models may be required for certain environments.
Compare options and pricing on the UtopikAI models page.

​
Where models are configured

Models can be selected and managed at several levels:
  • Tenant admin — Control which LLM and embedding models are available to a workspace, including filtering the model catalog.
  • Assistants — Pick the LLM (and related settings) when creating or editing an assistant. See Creating Assistants.
  • Studio — Choose generation models for images, video, audio, and 3D.
See Release Notes for the latest model and embedding improvements.

​
Pricing and usage

Usage costs depend on the models you select and how they are consumed (for example through chat and assistants versus API-style access). Workspace administrators can track consumption in Reporting. LLM pricing is on the LLM models page. Studio pricing is on the image, video, and audio pages. If you have questions about model selection or configuration, reach out to the UtopikAI team or support@utopikai.com.