Grok Build 0.1 vs Grok 4.7 vs Grok 4.3

Best cell per row highlighted. Null means undisclosed — never counted as zero.

xAI
Grok Build 0.1GA
6.7
AI Panel
xAI
Grok 4.7GA
8.2
AI Panel
xAI
Grok 4.3GA
7.7
AI Panel

Identity & lifecycle

ProviderxAIxAIxAI
Family / tierGrok Build · CoderGrok 4 · FlagshipGrok 4 · Flagship
StatusGAGAGA
Released2026-05-202026-09-212026-04-30
Knowledge cutoff2024-112026-052025-12

Architecture & context

Context window256K500K1M
Max output tokens256K——
Input modalitiestext, imagetext, imagetext, image, video
Reasoning modealwaysalwaysalways
Open weightsNoNoNo

Pricing (per Mtok)

Input$1$2$1.25
Output$2$6$2.5
Cached input$0.2$0.5$0.2
Batch input——$1
Free tierNoYesYes

Speed

Output speed (tok/s)——181.5
Time to first token (s)——19.69
Latency tierslowmediumslow

Trust & deployment

Trains on inputsNoNoNo
LicenseProprietaryProprietaryProprietary
Clouds——Azure AI Foundry
ComplianceSOC2SOC2SOC2

AI Panel scoring

Unified score6.78.27.7
Decision Maker6.587.5
Domain Strategist6.58.58
Finance Lead7.598.5
Domain Practitioner78.57.5
Power User6.587.5
Skeptic676.5
Value score7.58.88.5

Benchmarks

GPQA Diamond——90
SWE-bench Verified70.8——
IFEval——81
TAU-bench——98
LMArena Elo——1443
LMArena Coding Elo——1488

Scores link to their sources. Missing cells mean the vendor hasn't disclosed a result — honesty over padding.