Google Gemini

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.

Gemini composer showing mode selector dropdown with three options: Fast (answers quickly), Thinking (solves complex problems), and Pro (advanced math & code)

What's happening

Gemini abstracts technical model selection into user-friendly "modes" that emphasize outcomes rather than specifications. Instead of exposing model versions or parameter counts, Gemini presents Fast, Thinking, and Pro tiers that align with user goals (speed vs. depth).

Patterns

Model Selection UI

Dropdown presenting three performance modes with outcome-based descriptions

Open playground

Abstracted labels (Fast, Thinking, Pro) replace technical model names

UX Insights

•Mode-based framing makes AI complexity more approachable for general users
•Outcome descriptions ("Answers quickly", "Solves complex problems") guide selection by use case
•Three-tier structure provides clear upgrade path from free to premium
•Placement next to Tools icon creates a unified control center for customization
•Selected mode displays inline (e.g., "Fast" button) for at-a-glance awareness

Design Decisions

Gemini deliberately avoids exposing model version numbers or technical details, instead presenting selection as a spectrum from speed to power. This design choice prioritizes accessibility and outcome-based thinking over technical transparency. The mode selector lives in the composer rather than settings, suggesting Google treats it as a per-session preference rather than a global configuration.

More case studies

Claude

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.

DeepSeek

Tool Switching in Composer

DeepSeek composer: Instant/Expert tier toggle, DeepThink and Search chips in the card, and attach with tier-specific limits.

Grok

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.