← All models

Kimi K2.5

Moonshot AI

flagship

Capabilities

VisionToolsCodeExtended ThinkingStructured OutputMultilingualVideo

Strengths

  • Strong reasoning with thinking mode
  • Upgraded from K2
  • Competitive with frontier models
  • Video understanding (VideoMMMU 86.6%)
  • Agent swarm — parallel sub-task execution

Weaknesses

  • China-based
  • Smaller ecosystem

Pricing

Input / 1M tokens

$0.375

Output / 1M tokens

$2.025

Context window

128k

Transparency

Open weights

7.0 / 10

Open training data

3.0 / 10

Open methodology

5.0 / 10

Licence openness

7.0 / 10

Provider disclosure

2.0 / 10

FMTI company score

N/A

Composite:4.8 / 10

Not FMTI-scored.

Sustainability

Inference energy

5.0 / 10

Training footprint

N/A

Provider infrastructure

2.0 / 10

Composite:3.5 / 10

No provider energy data.

Task Fitness

question answering
80%
code
86%
mathematics
85%
business communication
64%
analysis
83%
extraction
81%
long-form generation
75%
research
58%
reasoning
83%
summarisation
78%
translation
78%
conversation
75%