HomeKimi-K2.5
1000B
Model parameter quantity
Moonshot
Affiliated organization
Open Source
License Type
January 27, 2026
Release time
Model Introduction
The latest model launched by Moonshot AI, featuring enhanced coding capabilities and superior performance in general Agent tasks, which adopts the version number kimi-k2.5.
Comprehensive score
Language dialogue
Knowledge reserve
Reasoning association
Mathematical calculation
Code writing
Command following
Swipe left and right to view more
Language comprehension ability
Capable of understanding complex contexts and generating logically coherent sentences, though occasionally off in tone control.
8.7
Knowledge coverage scope
Covers more than 200 specialized fields, integrating the latest research findings and cross-cultural knowledge in real time.
9.2
Reasoning ability
Unable to maintain coherent reasoning chains, often causing inverted causality or miscalculations.
5.4
Model comparison
Kimi-K2.5 vs Qwen2.5-7B-Instruct
Like Qwen2, the Qwen2.5 language models support up to 128K tokens and can generate up to 8K tokens. They also maintain multilingual support for over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, Arabic, and more.
Kimi-K2.5 vs Hunyuan-T1-20250822
The deep reasoning model independently developed by Tencent adopts the version number hunyuan-t1-20250822.
Kimi-K2.5 vs Spark-X1
The inference model Spark X1 released by iFlytek, on the basis of leading domestic mathematical tasks, benchmarks the performance of general tasks such as inference, text generation, and language understanding against OpenAI o1 and DeepSeek R1.
Related model
Kimi-K2-Thinking
The latest model launched by Moonshot AI, featuring enhanced coding capabilities and superior performance in general Agent tasks, which adopts the version number kimi-k2-thinking.
Kimi-K2-Instruct-0905
The latest model launched by Moonshot AI, featuring enhanced coding capabilities and superior performance in general Agent tasks, which adopts the version number kimi-k2-instruct-0905.
Kimi-K2-Instruct
The latest model launched by Moonshot AI, featuring enhanced coding capabilities and superior performance in general Agent tasks, built on an MoE architecture.
Moonshot-v1-32K
Moonshot-v1-16K is an API model produced by Moonshot, with the model version number being moonshot-v1-16k.
Moonshot-v1-32K
Moonshot-v1-16K is an API model produced by Moonshot, with the model version number being moonshot-v1-16k.
Relevant documents
DeepSeek Code poised for launch
As AI technology accelerates, DeepSeek is at a thrilling juncture. The AI company recently revealed it has secured over 70 billion yuan in funding. Leadership has emphasized a commitment to groundbreaking AI research over immediate commercial gains.
Musk’s Grok: 1.5 Trillion Parameters and Cursor Code Absorption—Game Changer or Bluff?
Elon Musk is finally making a move.In the AI programming race, OpenAI and Anthropic are accelerating, while xAI appears to be lagging. Musk has often stated his aim to rival Claude, yet despite multiple updates to the Grok4.X series, the results look
OpenAI Secretly Changes Charter to Make Removing Altman Harder
Following the 2023 coup-like incident, OpenAI has further solidified protections for CEO Sam Altman by updating its corporate bylaws. Recently released court documents reveal that Altman's position is now rock-solid, with substantially higher barrier
Meta AI now responds to buyer messages on Facebook Marketplace
Facebook Marketplace introduces new Meta AI features, including automated replies to buyer inquiries, the company announced Thursday. The platform also leverages AI to accelerate item listings, summarize seller profiles, and now lets sellers offer sh
OpenAI outlines AI economy with public wealth funds, robot taxes, and four-day week
As governments struggle to manage the economic impact of superintelligent machines, OpenAI has released a set of policy proposals outlining how wealth and work could be reshaped in an "intelligence age." The ideas blend traditional left-leaning mecha
