GoogleUnited States· released Jun 2025

Gemini 2.5 Pro

Name: Gemini 2.5 Pro
Author: Google

Ranked #36, 60.5% behind the frontier on average.

Google, United States. One of 14 models from this lab on the Index.

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

Snapshot September 8, 2026

OpenCharts Index

39.5

Ranked #36 of 36 · 60.5% behind the frontier on average · 18 comparable tests · 4 counting categories

Drop any single test and the Index lands between 35.2 and 41.9.

6 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

Gemini 2.5 Pro, in 18 results.

Strongest in coding (50.8) and math (36.8). It trails the frontier most in reasoning, 65.5% behind on average. No comparable results yet in knowledge and long context. Below 40% of the best published results, on average.

Strongest category

Coding

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

50.8

SWE-bench Verified57.6% · 69% of best
SciCode42.8% · 69% of best
WeirdML54.0% · 58% of best
Code Arena (WebDev)1,226 · 7% win rate vs leader

Mean of 4 of the category's 7 comparable tests. Counts toward the Index.

Best single result

GPQA Diamond

Answer a PhD-level multiple-choice question in physics, chemistry or biology that a web search will not settle.

85.3%

85.2% of the best (95.8%), above chance · #54 of 78.

Expert-level science questions are the closest thing to asking a specialist colleague. A model that gets them right can check a technical claim instead of echoing it.

Where it trails

FrontierMath Tier 4

Solve the hardest FrontierMath tier: problems that take expert mathematicians days.

0.0%

100.0% behind the best published result (97.6%) · #50 of 51.

The deepest end of the set. Progress here signals genuinely new capability.

Not measured yet

No comparable result yet in these 2 categories. A blank is a blank, never a zero, and it does not lower the Index.

Knowledge
Long context

Covered, but not averaged into the Index:

Multimodalsingle-test category, shown beside the Index · 81.1
Human preferencesingle-test category, shown beside the Index · 82.5

Every score, with its source

See it at work

What Gemini 2.5 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.

Human preference · mean 82.5 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.

1,446

41% expected win rate against the board leader (1,507)
#52 of 92

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

Gemini 2.5 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.

Gemini 2.5 Pro

#36·Index 39.5F

0 : 11

wins · 0 ties · 11 shared

GPT-6 Astra

#1·Index 99.0A+

GPT-6 Astra comes out ahead on 11 of the 11 tests both have taken.

Category means

Reasoning
34.5
99.6
Knowledge
—
100.0
Coding
50.8
97.7
Math
36.8
99.8
Agentic
35.7
98.5
Multimodal
81.1
—
Human preference
82.5
—

Shared tests · biggest gaps first

FrontierMath Tier 4Math
0.0%
97.6%
−100.0
ARC-AGI-2Reasoning
4.9%
95.0%
−94.9
Code Arena (WebDev)Coding
1,226
1,797
−92.8
CritPtReasoning
2.0%
31.7%
−92.0
APEX-AgentsAgentic
6.6%
46.7%
−84.6
FrontierMath (Tiers 1–3)Math
24.6%
93.7%
−73.8
ARC-AGI-1Reasoning
41.0%
98.5%
−58.4
WeirdMLCoding
54.0%
92.9%
−41.8
SciCodeCoding
42.8%
56.5%
−22.0
Mock AIME 2024–2025Math
84.2%
100.0%
−15.8
GPQA DiamondReasoning
85.3%
95.8%
−14.8

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

Reasoning34.5 · 4/6
Knowledge—
Coding50.8 · 4/7
Math36.8 · 3/5
Agentic35.7 · 5/8
Multimodalbeside81.1 · 1/1
Human preferencebeside82.5 · 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

85.3%

85.2% of the best (95.8%), above chance
31th percentile of 78 · n = 198

Gemini 2.5 Pro

Gemini 2.5 Pro, in 18 results.

What Gemini 2.5 Pro was asked to do, and how it did.

Gemini 2.5 Pro against whoever you pick.

Every test, every source

Compare with its neighbours.

The best models, ranked here, working inside Theo.