OpenAIUnited States· released Oct 2025

GPT-5 Pro

Name: GPT-5 Pro
Author: OpenAI

Provisional: 7 comparable tests so far. Ranked once it reaches 6 across 3 counting categories.

OpenAI, United States. One of 26 models from this lab on the Index.

Third-party publishedLow confidence (thin coverage)

Snapshot September 8, 2026

OpenCharts Index

58.5

Provisional: not enough comparable tests or counting categories to rank yet · 7 comparable tests · 1 counting categories

Drop any single test and the Index lands between 52.9 and 71.5.

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

GPT-5 Pro, in 7 results.

Strongest in reasoning (58.5). Its widest gap is on ARC-AGI-2, 80.7% behind the best published result. No comparable results yet in knowledge, agentic, multimodal, human preference and long context. Between 40% and 50% behind the best published results, on average.

Strongest category

Reasoning

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

58.5

SimpleBench61.6% · 75% of best
ARC-AGI-170.2% · 71% of best
Humanity's Last Exam31.6% · 68% of best
ARC-AGI-218.3% · 19% of best

Mean of 4 of the category's 6 comparable tests. Counts toward the Index.

Best single result

SimpleBench

Answer everyday trick questions about the physical and social world that most people find easy.

61.6%

75.2% of the best (81.9%) · #20 of 46.

Common sense about the real world is where fluent models still trip. People score in the eighties here.

Where it trails

ARC-AGI-2

Infer the hidden rule from a few input and output grid pairs, then apply it to a new grid.

18.3%

80.7% behind the best published result (95.0%) · #36 of 51.

Novel visual puzzles resist memorization, so they measure fluid intelligence rather than recall.

Not measured yet

No comparable result yet in these 5 categories. A blank is a blank, never a zero, and it does not lower the Index.

Knowledge
Agentic
Multimodal
Human preference
Long context

Covered, but not averaged into the Index:

Coding1 of 7 tests · needs 4 · mean 65.0
Math2 of 5 tests · needs 3 · mean 39.8

Every score, with its source

See it at work

What GPT-5 Pro 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.

Coding · mean 65.0 across 1 comparable test · not counted toward the Index yet (fewer than half the category's tests)

WeirdMLLeaderboard

Solve an unusual machine-learning task end to end: write the training code, run it, and hit an accuracy target.

60.4%

65.0% of the best (92.9%)
#31 of 69

2025-10-06 · high

Source: Håvard Ihle

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

GPT-5 Pro 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.

GPT-5 Pro

provisional·Index 58.5C-

0 : 5

wins · 0 ties · 5 shared

GPT-6 Astra

#1·Index 99.0A+

GPT-6 Astra comes out ahead on 5 of the 5 tests both have taken.

Category means

Reasoning
58.5
99.6
Knowledge
—
100.0
Coding
65.0
97.7
Math
39.8
99.8
Agentic
—
98.5

Shared tests · biggest gaps first

ARC-AGI-2Reasoning
18.3%
95.0%
−80.7
FrontierMath Tier 4Math
19.5%
97.6%
−80.0
FrontierMath (Tiers 1–3)Math
55.8%
93.7%
−40.4
WeirdMLCoding
60.4%
92.9%
−35.0
ARC-AGI-1Reasoning
70.2%
98.5%
−28.7

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

Reasoning58.5 · 4/6
Knowledge—
Coding65.0 · 1/7
Math39.8 · 2/5
Agentic—
Multimodal—
Human preference—
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.

Humanity's Last ExamLeaderboardReasoning

31.6%

68.0% of the best (46.5%)
53th percentile of 18 · n = 2,500

2025-10-06

Source: Epoch AI — AI Benchmarking Hub

ARC-AGI-2LeaderboardReasoning

18.3%

19.3% of the best (95.0%)
30th percentile of 51 · n = 120

2025-10-06

Source: Epoch AI — AI Benchmarking Hub

ARC-AGI-1LeaderboardReasoning

70.2%