Benchmark suiteUpdated May 27, 2026

GPT-5.2.

A model-level benchmark profile for GPT-5.2, combining Arena category ranks, suite indexes, commercial price and speed data, SWE-bench status, and lab context.

OpenAI lab page

Lab

OpenAI

Proprietary

Coverage

89%

8 of 9 tracked signals

Best Arena rank

#21

Document Arena

Best suite

#21

Coding / 71

Model profile

Benchmark signal map.

This view treats missing values as coverage gaps, then puts every available score, rank, cost, and speed signal into the same visual hierarchy.

Analyst view

GPT-5.2 is strongest on coding.

89% coverage

Its best composite result is rank #21 in the coding view, with a 71 coding index. The strongest raw Arena signal is Document Arena at rank #21 with 22,399 public votes.

Best current lane

Coding

#21 / 71 coding index

Strongest raw signal

Docs

#21, 1433 score

Commercial posture

$4.81

78.2s first token / 75 tok/s

Coverage gaps

Search Arena

Profile completeness

8 of 9 signals matched

89%

Arena categories

4/5

Docs, Code, Vision, Text

Commercial match

Live

GPT-5.2 (xhigh)

SWE-bench

bash-only

Raw identifier

gpt-5.2

Performance cockpit

Public benchmark scores

Arena rows show score, public rank, confidence interval, vote count, and publish date. Missing rows stay visible as coverage gaps.

Text Arena

Arena category

1,435#61

1431-1439 CI

46,492 votes

May 27, 2026

Arena public leaderboard dataset

WebDev Arena

Arena category

1,404#32

1388-1421 CI

1,458 votes

May 25, 2026

Arena public leaderboard dataset

Vision Arena

Arena category

1,232#35

1225-1239 CI

16,106 votes

May 18, 2026

Arena public leaderboard dataset

Search Arena

Arena category

Not listed

No confident public Arena row for this category.

Coverage gap

Document Arena

Arena category

1,433#21

1427-1439 CI

22,399 votes

May 12, 2026

Arena public leaderboard dataset

SWE-bench Verified

Software issue resolution

72.8%#9

bash-only

gpt-5-2

Feb 17, 2026

SWE-bench Verified

Cost and speed

Commercial profile

Price, latency, and throughput are only shown when a confident commercial source row is matched.

Blended price

$4.81

USD / 1M tokens

Input price

$1.75

USD / 1M input tokens

Output price

$14.00

USD / 1M output tokens

Latency

78.2s

Median time to first token

Output speed

75 tok/s

Median generation speed

Matched source row

GPT-5.2 (xhigh)

Suite position

Composite ranks by benchmark lens.

These views combine available public signals into purpose-specific indexes. A missing card means the model lacks enough source coverage for that lens.

Value

Quality, price, latency, and output speed

#7447

value index

blended / 1M tokens$4.81

time to first token78.2s

output speed75 tok/s

Coding

Code Arena, SWE-bench, and output speed

#2171

coding index

Code Arena rank #321,404