Skip to content

Thinking Modes

SUPERWISE Chat processes every message through one of four thinking modes. The mode you choose controls how deeply the AI reasons, how many sources it searches, and how long you wait for a response. It also affects how many tokens the message costs, shown as a multiplier below.

ModeSpeedDepthBest ForCost
Quick<2sStraightforward answers”What’s the date?” “Define X”0.5×
Thinking2–5sMulti-step reasoning”Pros and cons of Y?“1.0×
Research<15sSearch + reasoning”Competitor analysis on Z”1.5×
Thought Partner<30sDeep research + tools”Full impact assessment”2.0×

Pick a mode before you send a message. The default is Thinking.


Speed: Instant (<2 seconds) Process: Direct answer, single-pass search Best for: Factual questions, definitions, simple tasks

Example:

Q: "What's the capital of France?"
A: "Paris" (instant)

Speed: Fast (2–5 seconds) Process: Reasoning + search Best for: Multi-step problems, comparisons, analysis

Example:

Q: "Should we use tool A or B for X?"
A: [Weighs pros/cons, cites relevant docs, recommends A]

Speed: Medium (5–15 seconds, up to 30s) Process: Deep search + reasoning + synthesis Best for: Competitive analysis, trend analysis, complex decisions

Example:

Q: "Full competitive landscape for AI tools"
A: [Searches web + KB + docs, produces detailed matrix, cites 20+ sources]

Speed: Slow (10–30 seconds, up to 60s) Process: Full pipeline — search, reasoning, simulation, tools, forms Best for: High-stakes decisions, big picture strategy, detailed proposals

Example:

Q: "Should we hire 5 or 10 people this year?"
A: [Gathers budget/headcount data, runs simulations, produces decision matrix with financial impact]

  • Quick for facts, definitions, and one-liners — no point spending 4× the tokens on “What is SSO?”
  • Thinking for most everyday questions where you need reasoning, not just recall
  • Research when you need external sources, citations, or competitive information
  • Thought Partner for decisions with real consequences — budget calls, hiring plans, strategy reviews

The multiplier is the per-message token cost relative to Thinking (1.0×). Your monthly token allowance depends on your plan — see Plans & Pricing.