Benchmark suiteUpdated May 27, 2026

Kimi K2 0905 Preview.

A model-level benchmark profile for Kimi K2 0905 Preview, 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

44%

4 of 9 tracked signals

Best Arena rank

#83

Text Arena

Best suite

#15

Agents / 99

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 0905 Preview is strongest on agents.

44% coverage

Its best composite result is rank #15 in the agents view, with a 99 agent proxy. The strongest raw Arena signal is Text Arena at rank #83 with 11,795 public votes.

Best current lane

Agents

#15 / 99 agent proxy

Strongest raw signal

Text

#83, 1418 score

Commercial posture

$1.08

1.56s first token / 25 tok/s

Coverage gaps

WebDev ArenaVision ArenaSearch ArenaDocument ArenaSWE-bench

Profile completeness

4 of 9 signals matched

44%

Arena categories

1/5

Text

Commercial match

Live

Kimi K2 0905

SWE-bench

Not listed

No confident software benchmark row

Raw identifier

kimi-k2-0905-preview

Open 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

1,418#83

1412-1425 CI

11,795 votes

May 27, 2026

Arena public leaderboard dataset

WebDev Arena

Arena category

Not listed

No confident public Arena row for this category.

Coverage gap

Vision Arena

Arena category

Not listed

No confident public Arena row for this category.

Coverage gap

Search Arena

Arena category

Not listed

No confident public Arena row for this category.

Coverage gap

Document Arena

Arena category

Not listed

No confident public Arena row for this category.

Coverage gap

SWE-bench Verified

Software issue resolution

Not listed

No confident SWE-bench row matched to this model.

Coverage gap

Cost and speed

Commercial profile

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

Blended price

$1.08

USD / 1M tokens

Input price

$0.600

USD / 1M input tokens

Output price

$2.50

USD / 1M output tokens

Latency

1.56s

Median time to first token

Output speed

25 tok/s

Median generation speed

Matched source row

Kimi K2 0905

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

#8041

value index

blended / 1M tokens$1.08

time to first token1.56s

output speed25 tok/s

Coding

Code Arena, SWE-bench, and output speed

#1020

coding index

Code ArenaNot listed