VerdictPal
Join

Models / xAI

Grok 4.3 (medium)

Grok 4.3 · Released 2026-04-30

AA Intelligence Index 36.0. Listed on Artificial Analysis with API pricing $1.25/$2.5 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.

#85 of 184 on AA Index · current snapshot
25 AA Index · 8 benchmark rows · 1 sourcesPosition is benchmark-specific — not a cross-family or cross-source ranking.
Composite55
Context128K
Input / 1M$1.3
Output / 1M$2.5
Knowledge cutoffopen
Output speed137 t/s
TTFT (default API)17.59s
Statussolid

Family profile

Best published score in each covered benchmark family.

65/100 avg
Reasoning
89
Agentic
30
Long context
75

3 tested benchmark families

Benchmark placements

Where Grok 4.3 (medium) 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
IFBenchReasoning#1/ 12783Artificial Analysis2026-09-10
AA time to first tokenPerformance#14/ 190Artificial Analysis2026-09-10
AA output speedPerformance#20/ 190Artificial Analysis2026-09-10
GPQA DiamondReasoning#59/ 19089Artificial Analysis2026-09-10
Humanity's Last ExamReasoning#80/ 18530Artificial Analysis2026-09-10
AA-LCRLong context#82/ 18275Artificial Analysis2026-09-10
Artificial Analysis Intelligence IndexReasoning#90/ 18825Artificial Analysis2026-09-10
Terminal-BenchAgentic#158/ 18930Artificial Analysis2026-09-10
Family context

The three highest-scoring models with pages in each capability family. Where Grok 4.3 (medium) shows up, it's highlighted.

ReasoningGrok 4.3 (medium) · 89
  1. 1
    GPT-6 Astra
    96
  2. 2
    Claude Mythos Preview
    95
  3. 3
    Gemini 3.7 Flash
    95
AgenticGrok 4.3 (medium) · 30
  1. 1
    GPT 5.5
    100
  2. 2
    Claude Opus 4.8
    100
  3. 3
    Claude Fable 5
    100
Long contextGrok 4.3 (medium) · 75
  1. 1
    Kimi K3
    89
  2. 2
    Claude Fable 5.1
    85
  3. 3
    GPT-5.5 (high)
    84
Newest receipts

Benchmark rows added to the public ledger for Grok 4.3 (medium) in the last 120 days. Older rows live in the full table below.

8 recent
BenchmarkFamilyScoreSourceDays ago
AA-LCRLong context75%Artificial Analysis4
AA output speedPerformance137 t/sArtificial Analysis4
AA time to first tokenPerformance17.59sArtificial Analysis4
Artificial Analysis Intelligence IndexReasoning24.8%Artificial Analysis4
GPQA DiamondReasoning89%Artificial Analysis4
Humanity's Last ExamReasoning30%Artificial Analysis4
IFBenchReasoning83.33%Artificial Analysis4
Terminal-BenchAgentic30.3%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.

Reasoning89
Agentic30
Long context75
Price vs. performance
0255075100$0.10$0.30$1$3$10$30input USD / 1M tokens · log scale →← intelligence

Hover any dot for name, score, and input price. Keyboard: tab through the top twelve, or use the ranking below.

AnthropicOpenAIMistralAlibabaDeepSeekGoogleMoonshotMetaxAIMiniMaxZhipuXiaomiOtherMeituanCohereNVIDIA

Full benchmark ledger

Every catalog benchmark for Grok 4.3 (medium). Scores link to the original source; gaps mean no public row exists yet.

8 sourced rows

8 of 37 catalog benchmarks have a sourced row for Grok 4.3 (medium).

Reasoning

Agentic

Long context

Performance

Sources