AI model and tool discovery guide

← Back to models

xAI

Grok 4.20

xAI’s flagship Grok reasoning tier with a very large context window and native tool-oriented positioning.

Overall score

90

reasoninglong-contexttool-use

Official site Official docs Pricing Start shortlist

Context window

2M tokens

Speed

Balanced

Input pricing

$2 / 1M input tokens ($0.20 cached)

Output pricing

$6 / 1M output tokens

Score breakdown

Capability93

Use-case fit90

Cost efficiency82

Speed86

Reliability86

Agent readiness92

Ecosystem84

Scores combine benchmark signals, product experience, and editorial weighting. Use them as a practical guide, not an absolute truth claim.

Best for

ResearchAgent automation

Works with

xAI APIserver-side search toolsMCP-capable agent stacks

Modalities

textimage

Sources & trust

Officially verified core fields

Official linkSummaryDescriptionModalitiesModality profileContext windowTool supportPricingPricing page

Editorial fields such as shortlist guidance, strengths, caveats, and scoring remain clearly separated from official provider data.

Official site · Tier 5 · Apr 13, 2026

xAI models and pricing

Official docs · Tier 5 · Apr 13, 2026

SummaryDescriptionModalitiesModality profileContext windowTool supportPricingPricing page

Grok 4.20 VerdictLens review

Manual review · Tier 3 · Apr 9, 2026

Best-fit guidanceWorks-with guidanceStrengthsCaveatsOverall scoreScore breakdown

Last verified: Apr 13, 2026

Strengths

Big context window and official support for tool-centric workflows are the main draw.
Useful for teams wanting frontier-style reasoning outside the usual OpenAI/Anthropic duopoly.

Things to watch

xAI’s ecosystem depth is still thinner than the most mature platforms.
Official pricing is bundled with xAI-native server-side tools, so cost modeling needs care.

Best for

Research synthesis & analyst workflows

Prioritize source grounding, multilingual reading, long-context reasoning, and a retrieval stack that stays inspectable.

Agent automation & operations

Prioritize tool reliability, composability, secret handling, and robust state management across long-running flows.