DataLearner logoDataLearnerAI
Latest AI Insights
Model Leaderboards
Benchmarks
Model Directory
Model Comparison
Resource Center
Tools
LanguageEnglish
DataLearner logoDataLearner AI

A knowledge platform focused on LLM benchmarking, datasets, and practical instruction with continuously updated capability maps.

Products

  • Leaderboards
  • Model comparison
  • Datasets

Resources

  • Tutorials
  • Editorial
  • Tool directory

Company

  • About
  • Privacy policy
  • Data methodology
  • Contact

© 2026 DataLearner AI. DataLearner curates industry data and case studies so researchers, enterprises, and developers can rely on trustworthy intelligence.

Privacy policyTerms of service
Page navigation
Page navigation
Model catalogKimi K2.5
KI

Kimi K2.5

Multimodal model

Kimi K2.5

Release date: 2026-01-27Updated: 2026-03-08 21:06:20Knowledge cutoff: 2024-045,382
Live demoGitHubHugging FaceCompare
Parameters
1000B
Context length
256K
Chinese support
Supported
Reasoning ability

Kimi K2.5 is an AI model published by Moonshot AI, released on 2026-01-27, for Multimodal model, with 10000.0B parameters, and 256K tokens context length, requiring about 595GB storage, under the Modified MIT License license.

Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology

Kimi K2.5

Model basics

Reasoning traces
Supported
Thinking modes
Standard ModeThinking Level · Extended
Context length
256K tokens
Max output length
16384 tokens
Model type
Multimodal model
Release date
2026-01-27
Model file size
595GB
MoE architecture
Yes
Total params / Active params
1000B / 32B
Knowledge cutoff
2024-04
Kimi K2.5

Open source & experience

Code license
Modified MIT License
Weights license
Modified MIT License- 免费商用授权
GitHub repo
https://github.com/MoonshotAI/Kimi-K2
Hugging Face
https://huggingface.co/moonshotai/Kimi-K2.5
Live demo
https://www.kimi.com/en
Kimi K2.5

Official resources

Paper
Kimi K2.5: Visual Agentic Intelligence
DataLearnerAI blog
DataLearnerAI blog
Kimi K2.5

API details

API speed
2/5
💡Default unit: $/1M tokens. If vendors use other units, follow their published pricing.
Standard pricingStandard
ModalityInputOutput
Text$0.6$3
Image$0.6--
Cached pricingCache
ModalityInput cacheOutput cache
Text$0.1--
Image$0.1--
Kimi K2.5

Benchmark Results

Kimi K2.5 currently shows benchmark results led by HLE (17 / 149, score 50.20), LiveCodeBench (14 / 118, score 85), GPQA Diamond (31 / 175, score 87.60). This page also consolidates core specs, context limits, and API pricing so you can evaluate the model from benchmark results and deployment constraints together.

Thinking
Tool usage
Internet

General Knowledge

6 evaluations
Benchmark / mode
Score
Rank/total
GPQA Diamond
Thinking Mode
87.60
31 / 175
MMLU Pro
Thinking Mode
78.50
64 / 124
ARC-AGI
Thinking Mode
65.30
31 / 65
HLE
Thinking Mode
30.10
69 / 149
HLE
Thinking ModeTools
50.20
17 / 149
ARC-AGI-2
Thinking Mode
11.80
35 / 58

Coding and Software Engineer

4 evaluations
Benchmark / mode
Score
Rank/total
LiveCodeBench
Thinking Mode
85
14 / 118
SWE-bench Verified
Thinking ModeTools
76.80
22 / 103
SWE-bench Multilingual
Thinking Mode
73
8 / 17
SWE-Bench Pro - Public
Thinking ModeTools
50.70
25 / 36

Math and Reasoning

4 evaluations
Benchmark / mode
Score
Rank/total
AIME2025
Thinking Mode
96.10
21 / 106
AIME 2026
Thinking Mode
92.50
10 / 14
IMO-AnswerBench
Thinking Mode
81.80
12 / 17
FrontierMath - Tier 4
Standard Mode
4.20
40 / 80

Math and Reasoning

1 evaluations
Benchmark / mode
Score
Rank/total
Simple Bench
Thinking Mode
46.80
13 / 27

AI Agent - Information Search

1 evaluations
Benchmark / mode
Score
Rank/total
BrowseComp
Thinking ModeToolsInternet
60.60
27 / 43

AI Agent - Tool Usage

1 evaluations
Benchmark / mode
Score
Rank/total
Terminal Bench 2.0
Thinking ModeTools
50.80
30 / 43

Productivity Knowledge

1 evaluations
Benchmark / mode
Score
Rank/total
GDPval-AA
Thinking Mode
40
14 / 20

Long Context

1 evaluations
Benchmark / mode
Score
Rank/total
AA-LCR
Thinking Mode
65
10 / 13

Claw-style Agent Evaluation

2 evaluations
Benchmark / mode
Score
Rank/total
Pinch Bench
Thinking ModeTools
84.80
17 / 37
Claw Bench
Thinking ModeTools
81.70
18 / 29
View benchmark analysisCompare with other models

Compare with other models

  • Peer modelKimi K2.5 vs GLM-514 benchmarks
  • Peer modelKimi K2.5 vs MiniMax M2.513 benchmarks
  • Earlier versionKimi K2.5 vs Kimi K29 benchmarks
  • Earlier versionKimi K2.5 vs Kimi K2 Thinking9 benchmarks
  • Earlier versionKimi K2.5 vs Kimi K2 09054 benchmarks

Want a custom combination? Open the compare tool

Kimi K2.5

Publisher

Moonshot AI
Moonshot AI
View publisher details
Kimi K2.5

Model Overview

Kimi K2.5 is an AI model published by Moonshot AI, released on 2026-01-27, for Multimodal model, with 10000.0B parameters, and 256K tokens context length, requiring about 595GB storage, under the Modified MIT License license.

DataLearner on WeChat

Follow DataLearner on WeChat for AI model updates and research notes.

DataLearner WeChat QR code