Kimi
Kimi K2.5 (Reasoning)
kimi-k2.5 is Kimi's most versatile model to date, featuring a native multimodal architecture that supports both visual and text input, thinking and non-thinking modes, and dialogue and agent tasks.
Kimi model releasesModel Snapshot
TypeLanguage
ReleaseJan 26, 2026
Context262,144
Tagsreasoning, vision, tool-use, implicit-caching
Overall
Rank #110 across 600
35.4
Coding
Rank #109 across 266
46.8
Blended Price
Rank #206 across 379
$1.20/M
Strength Profile
Percentile score by analysis domain.
* Cost is inverted: lower input, output, and blended prices rank higher.
Benchmark Percentiles
Higher bars mean stronger relative placement.
Model Details
Runtime
Max Output262,144
Streaming speed is not measured for this model yet.
Pricing
Blended$1.20/M
Input$0.600/M
Output$3.00/M
Catalog$0.600/M in / $3.00/M out
Metadata
Modalities
All Benchmarks
| Metric | Domain | Value | Rank |
|---|---|---|---|
| Artificial Analysis Intelligence Index | overall | 35.4 | #110 |
| Artificial Analysis Coding Index | coding | 46.8 | #109 |
| GPQA | reasoning | 87.9% | #63 |
| Humanity's Last Exam | reasoning | 29.4% | #66 |
| SciCode |