xAIUnited States· released Sep 2026

Grok 4.7

Name: Grok 4.7
Author: xAI

Ranked #36, 40.9% behind the frontier on average.

xAI, United States. One of 9 models from this lab on the Index.

Third-party publishedHigh confidence (12+ comparable tests, 4+ Index categories)

Snapshot November 10, 2026

OpenCharts Index

59.1

Ranked #36 (27 to 39) of 80 · 40.9% behind the frontier on average · 24 comparable tests · 5 counting categories

Drop any single test and the Index lands between 55.6 and 62.5, anywhere from #27 to #39. Neighbours inside that range are ties.

4 of 5 hardest tests · share of the frontier

Distance to today's frontier on the hardest tests. Not a measure of intelligence.

What it's known for

Grok 4.7, in 24 results.

Strongest in coding (68.2) and agentic (64.7). It trails the frontier most in math, 50.8% behind on average. Between 40% and 50% behind the best published results, on average. Drop any single test and it would sit anywhere from #27 to #39.

Strongest category

Coding

Writing, fixing and shipping real code, judged by tests or users.

68.2

FrontierCode47.6% · 87% of best
SciCode57.8% · 86% of best
CursorBench46.3% · 80% of best
Code Arena (WebDev)1,639 · 53% win rate vs leader
WeirdML v318.7% · 35% of best

Mean of 5 of the category's 6 comparable tests. Counts toward the Index.

Best single result

Mock AIME 2024–2025

Solve competition mathematics problems that have a single exact integer answer.

98.1%

98.1% of the best (100.0%) · #28 of 98.

Olympiad-style problems need long, careful chains of reasoning, and there is no partial credit.

Where it trails

Furniture Assembly

Look at a photo of furniture partway through assembly, compare it with the manual, and say whether it has been built correctly so far.

20.8%

100.0% behind the best published result (90.0%), measured above chance · #29 of 37.

Matching a diagram to a real object is the skill behind repairing an appliance or checking an installation from a photo.

Not measured yet

Covered, but not averaged into the Index:

Knowledgesingle-test category, shown beside the Index · 59.6
Human preferencesingle-test category, shown beside the Index · 76.3

Every score, with its source

See it at work

What Grok 4.7 was asked to do, and how it did.

Every test it has taken, by category, with the kind of task it faced, who produced the number and where the result landed against the best published one. The examples are original and illustrative, never items from the datasets themselves.

Human preference · mean 76.3 across 1 comparable test · a single-test category, shown beside the Index and never averaged into it

Text ArenaLeaderboard

Answer the same prompt as a rival model; real users vote blind on which answer they prefer.

1,442

38% expected win rate against the board leader (1,526)
#70 of 113

xhigh · board 2026-11-09

Source: Arena

Every result is read against the best published one on its test: percent scores as a share of the best above chance, Arena ratings as the expected win rate against the board leader, open-ended values on a log scale. Sitting a harder exam never lowers a model. Every number links to the publisher that produced it.

Head to head

Grok 4.7 against whoever you pick.

Choose a rival. Every counted test both have taken appears side by side as a share of the best published result on that test, with the category means above. A win is a higher share; a gap under 2 points is a tie, because one item on a 45-problem exam is 2.2 points and publishers' own standard errors are of that order.

Grok 4.7

#36·Index 59.1C-

0 : 21

wins · 2 ties · 23 shared

GPT-6 Astra

#1·Index 93.7A

GPT-6 Astra comes out ahead on 21 of the 23 tests both have taken.

Category means

Reasoning
62.4
96.1
Knowledge
59.6
80.4
Coding
68.2
93.4
Math
49.2
98.8
Agentic
64.7
90.6
Multimodal
50.9
89.6
Human preference
76.3
85.2

Shared tests · biggest gaps first

Furniture AssemblyMultimodal
20.8%
80.0%
−83.3
FrontierMath Tier 4Math
17.1%
97.6%
−80.5
ProofBenchMath
26.0%
99.0%
−73.0
Mystery Game PuzzlesReasoning
29.0%
84.0%
−68.1
WeirdML v3Coding
18.7%
54.0%
−65.4
FrontierSWE v2Agentic
29.5%
65.5%
−51.5
Chess PuzzlesReasoning
41.0%
74.0%
−44.5
FrontierMath (Tiers 1–3)Math
53.0%
93.7%
−42.8
CritPtReasoning
18.0%
31.7%
−38.1
ARC-AGI-2Reasoning
61.4%
95.0%
−35.4
Code Arena (WebDev)Coding
1,639
1,767
−32.8
GDP.pdfMultimodal
22.8%
34.2%
−32.6
Blueprint-Bench 2Multimodal
32.5%
49.7%
−29.2
SimpleQA VerifiedKnowledge
56.0%
75.6%
−20.8
APEX-AgentsAgentic
54.6%
64.7%
−12.0
Vending-Bench 2Agentic
$10,537
$15,515
−11.1
FrontierCodeCoding
47.6%
53.3%
−10.4
Vision ArenaMultimodal
1,249
1,283
−9.7
Text ArenaHuman preference
1,442
1,474
−8.9
ARC-AGI-1Reasoning
90.2%
98.5%
−8.5
GPQA DiamondReasoning
92.7%
95.8%
−4.2
Mock AIME 2024–2025Math
98.1%
100.0%
tie
SciCodeCoding
57.8%
56.5%
tie

Bars are distances to the best published result on each test (percent scores above chance, Arena ratings as win rate against the leader, open-ended values on a log scale). Raw values as the publishers report them.

Open GPT-6 Astra

Scores

Every test, every source

The best published run per benchmark, the raw value as the publisher reports it, who produced the number, how close it comes to the best published result on that test, and where the model sits among every model scored on it. Percent scores are read above chance, Arena ratings as a win rate against the board leader, open-ended values on a log scale. A test with too few models to compare against is shown but not counted, and so is an archived test: one whose publisher no longer covers new models.

Category means

Reasoning62.4 · 6/8
Knowledgebeside59.6 · 1/1
Coding68.2 · 5/6
Math49.2 · 4/4
Agentic64.7 · 3/6
Multimodal50.9 · 4/4
Human preferencebeside76.3 · 1/1
Long context—

Mean · tests taken of the category's comparable tests. A faded category either rests on fewer than half of them or is a single-test category (marked “beside”), which is shown next to the Index and never averaged into it.

GPQA DiamondEpoch-runReasoning

92.7%

94.9% of the best (96.4%), above chance
76th percentile of 96 · n = 198

xhigh

Grok 4.7

Grok 4.7, in 24 results.

What Grok 4.7 was asked to do, and how it did.

Grok 4.7 against whoever you pick.

Every test, every source

Compare with its neighbours.

The best models, ranked here, working inside Theo.