LANGUAGE MODEL Google/DeepMind

Chain-of-Thought Prompting (CoT)

A prompting technique that simply asks the model to "show its work" — generate intermediate reasoning steps before the final answer — and produces large accuracy gains on math and reasoning problems. Free, requires no model retraining, and works only on sufficiently large models (~60B+ parameters).

Why it matters

CoT is the discovery that LLM capability isn't just a function of parameters and data — it's also a function of how you elicit the model's computation at inference time. This three-axis view (params × data × inference compute) is now the standard frame for LLM capability planning.

Core Capabilities

Long Documents
Handles entire codebases, books, and multi-doc RAG.
Research
Foundational paper or scientific contribution.

Context Window

Context window not disclosed.

Availability

API
Not available
Product / App
Not available
Open Source
Not released
Enterprise

Pricing Model

Research artifact
Not commercially released.
Research

Capability / Performance

Where this model sits relative to the middle 60% of models in the tree. All scores are 0–10 (higher is better).

Lower 20% Upper 80% This model
Lower 20% 20th percentile — 20% of models score below this This model Where the current model lands Upper 80% 80th percentile — only 20% of models score above this Percentile boundaries are computed across every model in the tree that reports the underlying benchmark for each capability.

What it feels like

  • Language model from Google — see the linked sources below for benchmark and review coverage

Best use cases

  • General-purpose tasks within Google's deployment footprint
  • See the model spec and sources block for benchmarked use cases

Tools to try

Not ideal for

  • Tasks far outside the modalities listed in this model's spec
  • Workflows where a more recent successor in the same family scores higher

Wei, J. · Wang, X. · Schuurmans, D. · Bosma, M. · Ichter, B. · Xia, F. · Chi, E. · Le, Q. · Zhou, D.