llmboard.ai
Benchmarks
CompareRankings
llmboard.ai
Benchmarks
CompareRankings
HomeModelsKimi K2.5

Moonshot AI model product

Kimi K2.5

Kimi K2.5 is Moonshot AI's flagship agentic model and a new SOTA open model. It unifies vision and text, thinking and non-thinking modes, and single-agent and multi-agent execution into one model. Built with Full-Parameter RL tuning, it achieves state-of-the-art performance across agents, coding, image, and video benchmarks.

Updated Aug 10, 2026. Default version: Kimi K2.5

Compare
LLMBoard score71.6Kimi K2.5
Coverage100%9 benchmark families
Context window262.1KTokens
Official input price$0.60Moonshot AI API

On this page

  • Specification
  • Capability
  • Benchmarks
  • Arena
  • Pricing
  • Versions
  • About
  • Compare
  • Similar models
  • FAQ

Model specification

Structured fields from the published default version.

Version
Kimi K2.5
Released
Jan 27, 2026
Knowledge cutoff
Jan 1, 2025
Parameters
1T
Context window
262.1K
Max output
262.1K
Inputs
image, text, video
Outputs
text
Open weights
Yes
License
MIT

Capability profile

This profile uses the latest version under this unique model that has a calculated LLMBoard score. Arena and price are excluded.

Kimi K2.5 category scores

Benchmark results

Published benchmark records for the scored version Kimi K2.5.

40 rows
Columns

Show columns

FinSearchComp T2&T30.711100.0%CAug 7, 2026
InfoVQAtest0.9112100.0%CAug 7, 2026
MathVista-Mini0.9123100.0%CAug 7, 2026
MMVU0.814100.0%CAug 7, 2026
OCRBench0.9122100.0%CAug 7, 2026
OJBench (C++)0.611100.0%CAug 7, 2026
Seal-00.616100.0%CAug 7, 2026
PaperBench0.62350.0%CAug 7, 2026
WideSearch0.82987.5%CAug 7, 2026
AA-LCR0.731585.7%CAug 7, 2026
LongVideoBench0.83433.3%CAug 7, 2026
MotionBench0.7330.0%CAug 7, 2026
VideoMMMU0.932692.0%CAug 7, 2026
ZEROBench0.14962.5%CAug 7, 2026
LVBench0.852482.6%CAug 7, 2026
SimpleVQA0.751366.7%CAug 7, 2026
Video-MME0.951775.0%CAug 7, 2026
WorldVQA0.5550.0%CAug 7, 2026
HMMT 20251.063384.4%CAug 7, 2026
LongBench v20.661768.8%CAug 7, 2026
DeepSearchQA0.87814.3%CAug 7, 2026
MMLU-Pro0.9712995.3%CAug 7, 2026
SciCode0.571864.7%CAug 7, 2026
LiveCodeBench v60.895384.6%CAug 7, 2026
OmniDocBench 1.50.991646.7%CAug 7, 2026
MathVision0.8103271.0%CAug 7, 2026
CyberGym0.411110.0%CAug 7, 2026
IMO-AnswerBench0.8131933.3%CAug 7, 2026
SWE-bench Multilingual0.7133463.6%CAug 7, 2026
FrontierSWE0.314157.1%BAug 7, 2026
MMMU-Pro0.8186573.4%CAug 7, 2026
AIME 20251.01911484.1%CAug 7, 2026
Humanity's Last Exam0.5209279.1%CAug 7, 2026
CharXiv-R0.8254747.8%CAug 7, 2026
BrowseComp0.7265856.1%CAug 7, 2026
SWE-Bench Verified0.82710474.8%CAug 7, 2026
LiveBench0.7323816.2%BAug 7, 2026
Terminal-Bench 2.00.5354929.2%CAug 7, 2026
GPQA0.93823384.0%CAug 7, 2026
SWE-Bench Pro0.542444.7%CAug 7, 2026

Arena results

Preference and agent-evaluation signals from published Arena datasets.

No published Arena match

The default version has no published Arena rows, or its source alias has not been resolved.

Pricing

Official vendor API PAYG pricing is summarized first. The table then lists individual provider offerings without treating their minimum as the official price.

Official API
$0.60 input, $3 output per 1M
Official provider
Moonshot AI
Lowest third-party
From $0.35 input, $1.7 output per 1M via CrofAI
Tracked offerings
23
22 rows
Columns

Show columns

Alibaba Coding Plan (China)kimi-k2.5globalN/AN/A262.1KAug 7, 2026
Alibaba Coding Plankimi-k2.5globalN/AN/A262.1KAug 7, 2026
Alibaba Token Plankimi-k2.5globalN/AN/A262.1KAug 7, 2026
Tencent Coding Plan (China)kimi-k2.5globalN/AN/A262.1KAug 7, 2026
Alibaba Token Plan (China)kimi-k2.5globalN/AN/A262.1KAug 7, 2026
CrofAIkimi-k2.5global$0.35$1.7262.1KAug 7, 2026
DigitalOceankimi-k2.5global$0.375$2.03262.1KAug 7, 2026
LLM Gatewaykimi-k2.5global$0.405$1.98262.1KAug 7, 2026
Cortecskimi-k2.5global$0.495$2.77262.1KAug 7, 2026
Aurikokimi-k2.5global$0.50$2.8262.1KAug 7, 2026
Charm Hyperkimi-k2.5global$0.55$2.89262.1KAug 7, 2026
Alibaba (China)kimi-k2.5global$0.574$2.41262.1KAug 7, 2026
OpenCode Gokimi-k2.5global$0.60$3262.1KAug 7, 2026
DaoXEkimi-k2.5global$0.60$3262.1KAug 7, 2026
OpenCode Zenkimi-k2.5global$0.60$3262.1KAug 7, 2026
FrogBotkimi-k2.5global$0.60$3256KAug 7, 2026
Abacuskimi-k2.5global$0.60$3262.1KAug 7, 2026
Azure Cognitive Serviceskimi-k2.5global$0.60$3262.1KAug 7, 2026
Moonshot AI (China)kimi-k2.5global$0.60$3262.1KAug 7, 2026
Azurekimi-k2.5global$0.60$3262.1KAug 7, 2026
AIHubMixkimi-k2.5global$0.60$3262.1KAug 7, 2026
Moonshot AIkimi-k2.5global$0.60$3262.1KAug 7, 2026

Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.

Kimi K2.5 versions

All published versions linked to this unique model. The score columns identify the version used by the current overall ranking.

1 rows
Columns

Show columns

Kimi K2.5Jan 27, 202671.61T262.1K262.1KYesMIT

What is Kimi K2.5?

A concise description based on the published model registry.

Kimi K2.5 is Moonshot AI's flagship agentic model and a new SOTA open model. It unifies vision and text, thinking and non-thinking modes, and single-agent and multi-agent execution into one model. Built with Full-Parameter RL tuning, it achieves state-of-the-art performance across agents, coding, image, and video benchmarks.

Use the benchmark, Arena and pricing sections above as separate evidence. A missing field means the current data snapshot does not support that claim.

Data snapshot: 2026-08-07. Editorial model content is not available in the backend.

Kimi K2.5 vs nearby models

Open a comparison with the three ranked models immediately above and below this model.

Kimi K2.5vsQwen3.6 27BKimi K2.5vsNemotron 3 UltraKimi K2.5vsDeepSeek-V4-FlashKimi K2.5vsGPT-5Kimi K2.5vsLlama 3.1 405BKimi K2.5vsDeepSeek-V2.5

Models similar to Kimi K2.5

Recommendations prioritize the same model type and family, then the closest published LLMBoard score.

#7+11.2
MA

Kimi K2.6

Moonshot AI

82.8 LLMBoard

DetailsCompare
#53-12.2
MA

Kimi K2 Thinking

Moonshot AI

59.4 LLMBoard

DetailsCompare
#80-20.2
MA

Kimi K2

Moonshot AI

51.4 LLMBoard

DetailsCompare
#30-0.0
OP

GPT-5

OpenAI

71.6 LLMBoard

DetailsCompare
#28+0.1
DE

DeepSeek-V4-Flash

DeepSeek

71.7 LLMBoard

DetailsCompare
#27+0.5
NV

Nemotron 3 Ultra

NVIDIA

72.1 LLMBoard

DetailsCompare

FAQ

Common questions about Kimi K2.5.

When was Kimi K2.5 released?

Kimi K2.5's default version was released on Jan 27, 2026.

How much does Kimi K2.5 cost?

Kimi K2.5's official API price is $0.60 per million input tokens and $3 per million output tokens via Moonshot AI. The lowest tracked third-party offer starts at $0.35 input and $1.7 output via CrofAI.

Who created Kimi K2.5?

Kimi K2.5 is published under Moonshot AI in the model registry.

What is the context window for Kimi K2.5?

The default version has a 262.1K token context window.

Is Kimi K2.5 open weight?

Yes. The default version is marked as open weight under MIT.

How many API providers offer Kimi K2.5?

23 published provider offerings are linked to the default version.

What models should I compare Kimi K2.5 with?

Nearby ranked alternatives include Qwen3.6 27B, Nemotron 3 Ultra, DeepSeek-V4-Flash.

Rankings

OverallCodingText ArenaPricing

Modalities

Image GenerationVideo GenerationSpeech-to-TextEmbeddings

Benchmarks

All BenchmarksReasoningMathCoding

Vendors

All VendorsOpenAIAnthropicGoogle
llmboard.aiCopyright 2026 llmboard.ai