Benchmark suiteLatest source data: Aug 30, 2026Checked: August 31, 2026

Kimi K2.5 Thinking.

A model-level benchmark profile for Kimi K2.5 Thinking, combining Arena category ranks, suite indexes, commercial price and speed data, SWE-bench status, and lab context.

Moonshot AI lab page

Lab

Moonshot AI

Open weights

Coverage

67%

6 of 9 tracked signals

Best Arena rank

#31

Document Arena

Best suite

#30

Coding / 59

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

Kimi K2.5 Thinking is strongest on coding.

67% coverage

Its best composite result is rank #30 in the coding view, with a 59 coding index. The strongest raw Arena signal is Document Arena at rank #31 with 19,891 public votes.

Best current lane

Coding

#30 / 59 coding index

Strongest raw signal

Docs

#31, 1435 score

Commercial posture

$1.20

Not listed first token / Not listed

Coverage gaps

Search ArenaLatencyOutput speed

Profile completeness

6 of 9 signals matched

67%

Arena categories

4/5

Docs, Vision, Code, Text

Commercial match

Live

Kimi K2.5 (Non-reasoning)

SWE-bench

#12

bash-only

Raw identifier

kimi-k2.5-thinking

ReasoningOpen weights

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

1,450#65

1447-1454 CI

70,610 votes

Aug 27, 2026

Arena public leaderboard dataset

WebDev Arena

Arena category

1,436#57

1431-1442 CI

19,044 votes

Aug 30, 2026

Arena public leaderboard dataset

Vision Arena

Arena category

1,250#43

1244-1256 CI

26,728 votes

Aug 27, 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

1,435#31

1428-1442 CI

19,891 votes

Jul 30, 2026

Arena public leaderboard dataset

SWE-bench Verified

Software issue resolution

70.8%#12

bash-only

kimi-k2.5

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

$1.20

USD / 1M tokens

Input price

$0.600

USD / 1M input tokens

Output price

$3.00

USD / 1M output tokens

Latency

Not listed

Median time to first token

Output speed

Not listed

Median generation speed

Matched source row

Kimi K2.5 (Non-reasoning)

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

#4459

value index

blended / 1M tokens$1.20

time to first tokenNot listed

output speedNot listed

Coding

Code Arena, SWE-bench, and output speed

#3059

coding index

Code Arena rank #571,436