← The Wire

Skills2026-06-07 · source-backed

Don't add chain-of-thought to reasoning models, give only the goal.

Confidence
source-backed
Sources
1
Redaction
passed

Story

On o3/o4-mini-class models with built-in reasoning, explicit CoT prompting gained only ~2.9 to 3.1% accuracy while adding 20 to 80% latency and cost. Net loss. Reserve CoT for non-reasoning models (it lifts Gemini Flash 2.0 ~13.5%), and for reasoning models supply only the objective and constraints. Source: Wharton GenAI Labs

Related stories

Each link below shares sources, entities, or timing with this story.

  1. Shared entity: CoT / Shared topic / What happened next / Tension

    Layer self-consistency on chain-of-thought for hard reasoning.

    Both cover CoT; overlapping topics (accuracy, chain-of-thought, model, reasoning); picks up the CoT thread on 2026-06-08.

  2. Shared entity: CoT / Shared topic / Earlier coverage / Downstream implication

    Thinking to Recall: Reasoning Unlocks Parametric Knowledge

    Both cover CoT; overlapping topics (cost, reasoning); earlier CoT coverage from 2026-03-11.

  3. Shared entity: Gemini Flash / Shared topic / Earlier coverage

    The $0.02/Call Coworker Pattern That Solves the Budget Problem

    Both cover Gemini Flash; overlapping topics (flash, gemini, model); earlier Gemini Flash coverage from 2026-05-03.

  4. Shared entity: CoT / Shared topic / Earlier coverage

    Reasoning Models Might Not Actually Reason. A New Paper Says They Decide Before They Think.

    Both cover CoT; overlapping topics (chain-of-thought, model, reasoning); earlier CoT coverage from 2026-04-02.

  5. Reasoning Theater: Performative Chain-of-Thought

    Both cover CoT; overlapping topics (chain-of-thought, model, reasoning); earlier CoT coverage from 2026-03-07.

  6. Shared entity: CoT / Shared topic / What happened next

    Chain-of-Symbol prompting beats Chain-of-Thought on spatial tasks

    Both cover CoT; overlapping topics (model, reasoning); picks up the CoT thread on 2026-07-02.

  7. Shared topic

    Gemini 3.2 Flash Leaked: 92% of GPT-5.5 Performance at 1/15th the Cost

    Overlapping topics (cost, flash, gemini, latency, model).

  8. Shared entity: CoT / Shared topic / Earlier coverage

    OpenAI Deploys GPT-5.4 Thinking to Monitor Its Own Coding Agents for Misalignment — In Production

    Both cover CoT; overlapping topics (chain-of-thought, reasoning); earlier CoT coverage from 2026-03-20.

Source trail

Entities

Provenance

AI generated
no
Story unit
2026-06-07-don-t-add-chain-of-thought-to-reasoning-models-give-only-the-goal
Labels
source-backed, canonical briefing excerpt
Don't add chain-of-thought to reasoning models, give only the goal. | MindPattern