Worth knowing: Which underlying AI model actually produced a given answer or image is often unclear — the interface labels modes like Smart/Genius/Creative rather than naming the model.
Summarized from 3 community tutorials · July 2026
Highlights
Multiple task-specific modesSmart, Genius, Research, and Create modes each tuned for a different kind of query — quick chat, multi-step reasoning, deep research, or image generation.
Switch between many underlying LLMsRe-run the same prompt across GPT-4o, Claude, Llama, Mistral and other models from one interface without leaving the chat.
Cited web research reportsResearch mode crawls dozens of sources (14+ sites in one example) and compiles a sourced list or report instead of raw chat text.
Built-in text-to-image generationCreate/Creative mode turns a text prompt into an image directly in the chat, using a stable-diffusion-based model.
Multi-step reasoning (Genius mode)Breaks a compound question into research + calculation steps (e.g., researching prices then computing savings) before giving a final answer.
Summarized from 3 community tutorials · July 2026
i
No approved reviews yet. Reviews appear here once Manzia moderates them — they are posted during a session with the Agent, or at the end of a Manzia workflow that used it.
