moonshotai

MoonshotAI: Kimi K2.5

MoonshotAI's Kimi K2.5 is designed for handling both text and image inputs with a substantial context length of 262144 tokens, allowing it to maintain coherence in lengthy conversations or tasks. The model supports reasoning and can work with tools, making it versatile for complex problem-solving scenarios. For users looking for a reliable performer at a moderate price point-pricing is $0.57 per million input tokens and $2.85 per million output tokens-the Kimi K2.5 stands out with its blended benchmark score of 73.3 across two independent evaluations, making it a solid choice despite the lack of structured output support in benchmarks.

Quality Score
100/100
price + capability + benchmarks
Input Price
$0.57
per 1M tokens
Output Price
$2.85
per 1M tokens
Context Window
262,144
tokens

Benchmark results

Independent, published benchmarks. Blended score 64.7 across 4 benchmarks, last refreshed 2026-07-27. How scoring works →

BenchmarkMeasuresScore
AI Index broad capability composite 58.4
AI Index Coding software engineering tasks 77.2
AI Index Agentic multi-step tool-using tasks 35.8
EQ-Bench emotional understanding in dialogue 72.5
Model ID
moonshotai/kimi-k2.5
Vendor
moonshotai
Released
January 2026
Tokenizer
Other
Input Modalities
text, image
Output Modalities
text
Max Output
262,144 tokens
Tool Calling
✓ supported
Structured Output
✓ supported
Reasoning Mode
✓ supported
Vision
✓ accepts images
Audio
no
Moderated
no

What it costs in practice

Computed from the current $0.57/M input and $2.85/M output rates. Run your own numbers →

JobTokensCost
Summarize a 50-page report 30k in / 1.5k out $0.02
Classify 1,000 customer emails 500k in / 50k out $0.43
A month of a busy support chatbot 5M in / 2M out $8.55

Category rankings

Where MoonshotAI: Kimi K2.5 places across the 26 categories it ranks in. How we rank →

#CategoryScore
#6 Short-Form SummarizationWriting · of 25 ranked 130
#6 Email DraftingWriting · of 25 ranked 126
#6 Language LearningEducation · of 25 ranked 126
#6 Chat CompanionPersonal · of 25 ranked 130
#6 Trivia & General KnowledgePersonal · of 25 ranked 120
#10 Image CaptioningVision · of 25 ranked 120
#13 Sales / Cold EmailBusiness · of 25 ranked 122
#13 Job Application DraftingBusiness · of 25 ranked 122
#13 Journaling HelperPersonal · of 25 ranked 122
#13 Recipe GenerationPersonal · of 25 ranked 122
#17 Resume WritingBusiness · of 25 ranked 116
#18 JSON ExtractionData · of 25 ranked 144
#19 Regex WritingCode · of 25 ranked 133
#19 Marketing CopyWriting · of 25 ranked 127
#19 Language TranslationWriting · of 25 ranked 133
#19 Standardized Test PrepEducation · of 25 ranked 127
#19 Tutoring for KidsPersonal · of 25 ranked 133
#19 Dataset AnnotationResearch · of 25 ranked 141
#20 Customer SupportBusiness · of 25 ranked 131
#21 Frontend Component DesignCode · of 25 ranked 141
#21 Screenshot DebuggingVision · of 25 ranked 138
#22 Chart & Graph ReadingVision · of 25 ranked 146
#23 Bulk Data LabelingData · of 25 ranked 133
#23 Blog Post WritingWriting · of 25 ranked 136
#23 Trip PlanningPersonal · of 25 ranked 136
#25 Diagram ExtractionVision · of 25 ranked 144

Similar models

Quick answers

How much does MoonshotAI: Kimi K2.5 cost?
$0.57 per million input tokens and $2.85 per million output tokens.
What is MoonshotAI: Kimi K2.5's context window?
262,144 tokens, roughly 393 pages of text in a single request.
Does MoonshotAI: Kimi K2.5 support tool calling?
It supports tool calling, structured output, a reasoning mode.
Can MoonshotAI: Kimi K2.5 process images?
Yes, it accepts image input.