xAIUnited States· released Feb 2026

Grok 4.20

Name: Grok 4.20
Author: xAI

Provisional: 14 comparable tests so far. Ranked once it reaches 6 across 3 counting categories.

xAI, United States. One of 8 models from this lab on the Index.

Third-party publishedLow confidence (thin coverage)

Snapshot September 8, 2026

OpenCharts Index

63.2

Provisional: not enough comparable tests or counting categories to rank yet · 14 comparable tests · 2 counting categories

Drop any single test and the Index lands between 42.9 and 68.0.

4 of 8 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 4.20, in 14 results.

Strongest in reasoning (83.5) and math (42.9). It trails the frontier most in math, 57.1% behind on average. No comparable result yet in long context. Between 35% and 40% behind the best published results, on average.

Strongest category

Reasoning

Hard, novel problems: graduate science, abstract puzzles, expert exams.

83.5

GPQA Diamond89.3% · 91% of best
ARC-AGI-189.5% · 91% of best
ARC-AGI-265.1% · 69% of best

Mean of 3 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.

92.2%

92.2% of the best (100.0%) · #33 of 78.

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

Where it trails

ProofBench

Write a complete, rigorous proof, graded for rigor rather than for a final answer.

14.0%

86.0% behind the best published result (100.0%) · #46 of 61.

A proof shows the reasoning itself. A right answer with a wrong argument scores nothing.

Not measured yet

No comparable result yet in this category. A blank is a blank, never a zero, and it does not lower the Index.

Long context

Covered, but not averaged into the Index:

Knowledgesingle-test category, shown beside the Index · 39.9
Coding2 of 7 tests · needs 4 · mean 36.2
Agentic2 of 8 tests · needs 4 · mean 69.8
Multimodalsingle-test category, shown beside the Index · 83.7
Human preferencesingle-test category, shown beside the Index · 90.6

Every score, with its source

See it at work

What Grok 4.20 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 90.6 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.

1,475

45% expected win rate against the board leader (1,507)
#22 of 92

beta1 · board 2026-09-02

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 4.20 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 4.20

provisional·Index 63.2C

0 : 10

wins · 0 ties · 10 shared

GPT-6 Astra

#1·Index 99.0A+

GPT-6 Astra comes out ahead on 10 of the 10 tests both have taken.

Category means

Reasoning
83.5
99.6
Knowledge
39.9
100.0
Coding
36.2
97.7
Math
42.9
99.8
Agentic
69.8
98.5
Multimodal
83.7
—
Human preference
90.6
—

Shared tests · biggest gaps first

ProofBenchMath
14.0%
99.0%
−85.0
Code Arena (WebDev)Coding
1,374
1,797
−83.9
FrontierMath Tier 4Math
17.1%
97.6%
−82.5
SimpleQA VerifiedKnowledge
30.2%
75.6%
−60.1
FrontierMath (Tiers 1–3)Math
44.9%
93.7%
−52.1
WeirdMLCoding
52.3%
92.9%
−43.7
ARC-AGI-2Reasoning
65.1%
95.0%
−31.4
GPQA DiamondReasoning
89.3%
95.8%
−9.1
ARC-AGI-1Reasoning
89.5%
98.5%
−9.1
Mock AIME 2024–2025Math
92.2%
100.0%
−7.8

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.

Category means

Reasoning83.5 · 3/6
Knowledgebeside39.9 · 1/1
Coding36.2 · 2/7
Math42.9 · 4/5
Agentic69.8 · 2/8
Multimodalbeside83.7 · 1/1
Human preferencebeside90.6 · 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

89.3%

90.9% of the best (95.8%), above chance
57th percentile of 78 · n = 198

reasoning

Source: Epoch AI — AI Benchmarking Hub

ARC-AGI-2LeaderboardReasoning

65.1%

68.6% of the best (95.0%)
64th percentile of 51 · n = 120

Source: Epoch AI — AI Benchmarking Hub

ARC-AGI-1LeaderboardReasoning

89.5%

90.9% of the best (98.5%)
58th percentile of 51 · n = 100