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.
Lab
Moonshot AI
Open weights
Coverage
56%
5 of 9 tracked signals
Best Arena rank
#37
Document Arena
Best suite
#56
Value / 52
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 value.
Its best composite result is rank #56 in the value view, with a 52 value index. The strongest raw Arena signal is Document Arena at rank #37 with 21,569 public votes.
Best current lane
Value
#56 / 52 value index
Strongest raw signal
Docs
#37, 1436 score
Commercial posture
$1.20
Not listed first token / Not listed
Coverage gaps
Profile completeness
5 of 9 signals matched
Arena categories
4/5
Docs, Vision, Code, Text
Commercial match
Live
Kimi K2.5 (Non-reasoning)
SWE-bench
Not listed
No confident software benchmark row
Raw identifier
kimi-k2.5-thinking
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
1447-1453 CI
WebDev Arena
Arena category
1431-1442 CI
Vision Arena
Arena category
1246-1258 CI
Search Arena
Arena category
Not listed
No confident public Arena row for this category.
Coverage gap
Document Arena
Arena category
1429-1443 CI
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
$1.20
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
value index
Coding
Code Arena, SWE-bench, and output speed
coding index
Agents
Search, documents, code, and latency
agent proxy
Reasoning
Text, documents, vision, and mode signal
reasoning proxy
Long context
Document, search, and text benchmark proxy
context proxy
Lab context
Nearby Moonshot AI profiles.
Same-lab models make it easier to see whether this profile is a flagship, value play, specialist, or lower-coverage row.