VerdictPal
Join

Models / xAI

Grok 4

Grok 4 · Released 2025-07-10

AA Intelligence Index 33.3. Listed on Artificial Analysis with API pricing $3/$15 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.

Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.

#86 of 131 on AA Index · current snapshot
33 AA Index · 14 benchmark rows · 1 sourcesPosition is benchmark-specific — not a cross-family or cross-source ranking.
Composite64
ContextUnlisted
Input / 1M$3.0
Output / 1M$15
Knowledge cutoffunpublished
Output speed0 t/s
TTFT (default API)0.00s
Statussolid
KnowledgeReasoningMathCodingAgenticLong context
Knowledge
87
Reasoning
88
Math
99
Coding
82
Agentic
75
Long context
68

83average across 6 tested families

Benchmark placements

Where Grok 4 places on each public benchmark source that publishes a row. Rank counts every model with a latest row in the same test — not a universal quality score.

BenchmarkFamilyRankScoreSourceDate
MATHMath#4/ 2499Artificial Analysis2026-07-28
AIMEMath#12/ 3293Artificial Analysis2026-07-28
LiveCodeBenchCoding#13/ 2682Artificial Analysis2026-07-28
MMLU-ProKnowledge#13/ 3487Artificial Analysis2026-07-28
τ³-Bench BankingAgentic#40/ 14375Artificial Analysis2026-07-28
AA-LCRLong context#43/ 14368Artificial Analysis2026-07-28
GPQA DiamondReasoning#52/ 14988Artificial Analysis2026-07-28
SciCodeCoding#63/ 14446Artificial Analysis2026-07-28
Humanity's Last ExamReasoning#79/ 14424Artificial Analysis2026-07-28
IFBenchReasoning#89/ 12154Artificial Analysis2026-07-28
Terminal-BenchAgentic#98/ 14838Artificial Analysis2026-07-28
Artificial Analysis Intelligence IndexReasoning#100/ 14733Artificial Analysis2026-07-28
AA output speedPerformance#121/ 149Artificial Analysis2026-07-28
AA time to first tokenPerformance#121/ 149Artificial Analysis2026-07-28
Family context

The three highest-scoring carded models in each capability family. Where Grok 4 shows up, it's highlighted.

KnowledgeGrok 4 · 87
  1. 1
    GPT 5.5
    91
  2. 2
    Claude Opus 4.8
    91
  3. 3
    Claude Sonnet 4.6
    91
ReasoningGrok 4 · 88
  1. 1
    Claude Mythos Preview
    95
  2. 2
    GPT-5.6 Sol
    94
  3. 3
    Kimi K3
    94
MathGrok 4 · 99
  1. 1
    OpenAI o3
    99
  2. 2
    GPT-5 (medium)
    99
  3. 3
    Grok 4
    99
CodingGrok 4 · 82
  1. 1
    GPT 5.5
    96
  2. 2
    Claude Opus 4.8
    96
  3. 3
    Claude Opus 4.6
    96
AgenticGrok 4 · 75
  1. 1
    GPT 5.5
    100
  2. 2
    Claude Opus 4.8
    100
  3. 3
    Claude Fable 5
    100
Long contextGrok 4 · 68
  1. 1
    GPT-5 (high)
    76
  2. 2
    GPT-5.2 Codex (xhigh)
    76
  3. 3
    Kimi K3
    75
Newest receipts

Benchmark rows added to the public ledger for Grok 4 in the last 120 days. Older rows live in the full table below.

14 recent
BenchmarkFamilyScoreSourceDays ago
AA-LCRLong context68%Artificial Analysis4
AA output speedPerformance0 t/sArtificial Analysis4
AA time to first tokenPerformance0.00sArtificial Analysis4
AIMEMath92.67%Artificial Analysis4
Artificial Analysis Intelligence IndexReasoning33.3%Artificial Analysis4
GPQA DiamondReasoning87.7%Artificial Analysis4
Humanity's Last ExamReasoning23.9%Artificial Analysis4
IFBenchReasoning53.67%Artificial Analysis4
LiveCodeBenchCoding81.9%Artificial Analysis4
MATHMath99%Artificial Analysis4
MMLU-ProKnowledge86.6%Artificial Analysis4
SciCodeCoding45.7%Artificial Analysis4
τ³-Bench BankingAgentic74.85%Artificial Analysis4
Terminal-BenchAgentic37.88%Artificial Analysis4
Family coverage

A high composite that hides a weak family is a trap. These bars surface the families where this model hasn't been publicly tested, and where it leads.

Knowledge87
Reasoning88
Math99
Coding82
Agentic75
Long context68
Price vs. performance
0255075100$0.10$0.30$1$3$10$30input USD / 1M tokens · log scale →← intelligence

Hover or tab any dot for name, score, and input price.

AnthropicOpenAIMistralAlibabaDeepSeekGoogleMoonshotxAIMiniMaxMetaZhipuXiaomiOtherMeituanCohereNVIDIA

Full benchmark ledger

Every catalog benchmark for Grok 4. Scores link to the original source; gaps mean no public row exists yet.

14 sourced rows

14 of 32 catalog benchmarks have a sourced row for Grok 4.

Knowledge

Reasoning

Math

Coding

Agentic

Long context

Performance

Sources