Model Selection
Grok offers intuitive model selection with Auto mode that intelligently chooses between Fast and Expert models, plus additional specialized models including Grok 4.1 Thinking, Heavy, and premium SuperGrok.
.png)
Six-tier model system from Auto to SuperGrok uses outcome-based labels (Fast, Expert, Heavy) instead of technical specs, making capabilities understandable at a glance
What's happening
Grok implements a tiered model selection strategy with Auto mode as the default, intelligently routing between Fast (4.1) and Expert models. The dropdown provides clear descriptions of each model: Auto (Chooses Fast or Expert), Fast (Quick responses by 4.1), Expert (Thinks hard), Grok 4.1 Thinking (Thinks fast), Heavy (Team of experts), and SuperGrok (Unlock extended capabilities - requires sign-in).
Patterns
Model selector with icon and dropdown showing model tiers with descriptions
Each model shows name and brief capability description
Auto mode intelligently chooses between Fast and Expert based on query
UX Insights
- •Auto mode as default removes decision burden from users
- •Six model tiers from Auto to SuperGrok provide clear upgrade path
- •Each model includes plain-language capability description
- •Model names use intuitive labels (Fast, Expert, Heavy) not technical specs
- •SuperGrok tier gated behind sign-in for premium access
- •Active model shown with checkmark and displayed next to composer
- •Icon-based visual system (lightning for Fast, lightbulb for Expert, grid for Heavy)
Design Decisions
Grok's model selection prioritizes intelligent defaults (Auto mode) while maintaining full transparency about available options. The tiered approach creates clear upgrade paths from speed (Fast) → depth (Expert) → specialized reasoning (Thinking) → collaborative (Heavy) → premium (SuperGrok). Descriptive labels like "Chooses Fast or Expert," "Quick responses by 4.1," and "Team of experts" communicate capabilities without technical jargon. The SuperGrok tier with sign-in requirement establishes clear freemium monetization.
More from the gallery

Model Switcher with Hierarchy
Claude implements a tiered model selection system that balances simplicity with power user needs. The main dropdown presents three current models with clear use-case descriptions, while a "More models" submenu provides access to legacy versions.

Personalization
ChatGPT provides a dedicated Personalization interface that allows users to customize how the AI behaves, including conversation style, response format, and contextual preferences, separate from technical settings.

Mode-Based Selection
Gemini reframes model selection as "mode" selection, presenting users with three performance tiers, Fast, Thinking, and Pro, each optimized for different task types and response complexity.

Model Selection Dropdown
Perplexity offers a comprehensive model switcher directly in the composer, displaying a wide array of AI models including GPT-5.2, Claude Opus 4.5, Gemini 3 Pro/Flash, Grok 4.1, and proprietary Sonar models.
More real-world AI UX in your inbox
Weekly gallery picks, interface patterns, and notes on how products ship AI - no spam, unsubscribe anytime.