Model

Grok 4.20

CLOSED

SpaceXAI · Grok 4.20 family · released Mar 10, 2026

xAI's March 2026 flagship reasoning model, sitting chronologically between Grok 4 Heavy (Jul 2025) and Grok 4.3 (Apr 2026). Public beta launched Feb 17, 2026 as 'Grok 4.20 (Beta)'; full Enterprise API GA on Mar 10, 2026. Official model slug (grok-4.20-0309-reasoning) and every documented alias use '4.20' — press/resellers sometimes informally shorten it to 'Grok 4.2', but xAI's own docs never use that spelling.

ReasoningCodingVisionFunction callingTool useAgentic

2736.7

Elo · unrated

Parameters

Undisclosed

Active params

Undisclosed

Context

1M tokens

Architecture

Mixture-of-Experts (size undisclosed); single-model reasoning configuration with native tool use, sibling to the 4-agent 'Grok 4.20 Multi-Agent' variant

License

Proprietary

Languages

—

API price (in/out)

$1.25 / $2.5

Modalities

text · vision

Benchmark results

Bar shows position within the tracked field; marker = field best

Arena EloHuman preference1474#3

best: Claude Fable 5 · 1505

IFBenchReasoning82.9%#2

best: MiniMax M3 · 83.0%

Run it locally

Closed weights — available via API only. No local deployment.

Input / M tok

$1.25

Output / M tok

$2.5

Grok 4.20 family

Elo progression across releases

Grok 4.20Mar 20262736.7 Grok 4.20 Multi-AgentMar 20261000.0

Resources

DOCShttps://docs.x.ai/developers/models/grok-4.20-reasoning APIAvailable via SpaceXAI API

API price $1.25/$2.5 · each benchmark row carries its own source badge (see methodology)