newsence
來源篩選

China's Moonshot AI Releases New Open-Source Multimodal Model Kimi K2.5 and a Coding Agent

Techcrunch

China's Moonshot AI has launched Kimi K2.5, a new open-source multimodal model capable of understanding text, images, and video. The model, trained on 15 trillion tokens, demonstrates strong performance in coding tasks and agent swarms, outperforming proprietary models in various benchmarks.

newsence

中國Moonshot AI發布全新開源多模態模型Kimi K2.5及程式碼代理

Techcrunch
大約 1 個月前

AI 生成摘要

中國Moonshot AI發布了Kimi K2.5,這是一個全新的開源多模態模型,能夠理解文字、圖像和影片。該模型經過15兆個token的訓練,在程式碼任務和代理協作方面表現出色,在多項基準測試中超越了專有模型。

China's Moonshot releases a new open-source model Kimi K2.5 and a coding agent | TechCrunch

Image Image

Topics

Latest

AI

Amazon

Apps

Biotech & Health

Climate

Cloud Computing

Commerce

Crypto

Enterprise

EVs

Fintech

Fundraising

Gadgets

Gaming

Google

Government & Policy

Hardware

Instagram

Layoffs

Media & Entertainment

Meta

Microsoft

Privacy

Robotics

Security

Social

Space

Startups

TikTok

Transportation

Venture

More from TechCrunch

Staff

Events

Startup Battlefield

StrictlyVC

Newsletters

Podcasts

Videos

Partner Content

TechCrunch Brand Studio

Crunchboard

Contact Us

Image

China’s Moonshot releases a new open-source model Kimi K2.5 and a coding agent

China’s Moonshot AI, which is backed by the likes of Alibaba and HongShan (formerly Sequioa China), today released a new open-source model, Kimi K2.5, which understands text, image, and video.

The company said that the model was trained on 15 trillion mixed visual and text tokens, and that’s why it is natively multimodal. It added that the models are good at coding tasks and handling agent swarms — an orchestration where multiple agents work together. In released benchmarks, the model matches the performance of the proprietary peers and even beats them in certain tasks.

For instance, in the coding benchmark, the Kimi K2.5 outperforms Gemini 3 Pro at the SWE-Bench Verified benchmark, and scores higher than GPT 5.2 and Gemini 3 Pro on the SWE-Bench Multilingual benchmark. In video understanding, it beats GPT 5.2 and Claude Opus 4.5 on VideoMMMU (Video Massive Multi-discipline Multimodal Understanding), a benchmark that measures how a model reasons over videos.

Image

Moonshot AI said that on the coding front, while the model can understand text well, users can also feed it images or videos and ask it to make a similar interface shown in those media files.

Here's a short video from our founder, Zhilin Yang.(It's his first time speaking on camera like this, and he really wanted to share Kimi K2.5 with you!) pic.twitter.com/2uDSOjCjly

To let people use these coding capabilities, the company has launched an open-source coding tool called Kimi Code, which would rival Anthropic’s Claude Code or Google’s Gemini CLI. Developers can use Kimi Code through their terminals or integrate it with development software such as VSCode, Cursor, and Zed. The startup said that developers can use images and videos as input with Kimi Code.

Coding tools have gained rapid popularity and are becoming revenue drivers for AI labs. Anthropic announced in November that Claude Code had reached $1 billion in annualized recurring revenue (ARR). Earlier this month, Wired reported that by the end of 2025, the tool had added $100 million to that figure. Moonshot’s Chinese competitor, Deepseek, is set to release a new model with strong coding chops next month, according to a report by The Information.

Moonshot was founded by former Google and Meta AI researcher Yang Zhilin. The company raised $1 billion in funding in a Series B round at a $2.5 billion valuation. According to Bloomberg, the startup picked up $500 million in funding last month at $4.3 billion valuation. What’s more, the report noted that it is already seeking to raise a new round at a $5 billion valuation.

Topics

Image

Ivan covers global consumer tech developments at TechCrunch. He is based out of India and has previously worked at publications including Huffington Post and The Next Web.

You can contact or verify outreach from Ivan by emailing [email protected] or via encrypted message at ivan.42 on Signal.

Image

Tickets are live at the lowest rates of the year. Save up to $680 on your pass — and if you’re among the first 500 registrants, score a +1 pass at 50% off.Meet investors. Discover your next portfolio company. Hear from 250+ tech leaders, dive into 200+ sessions, and explore 300+ startups building what’s next. Don’t miss these one-time savings.

Most Popular

Anthropic launches interactive Claude apps, including Slack and other workplace tools

Anthropic launches interactive Claude apps, including Slack and other workplace tools

TikTok users freak out over app’s ‘immigration status’ collection — here’s what it means

TikTok users freak out over app’s ‘immigration status’ collection — here’s what it means

Researchers say Russian government hackers were behind attempted Poland power outage

Researchers say Russian government hackers were behind attempted Poland power outage

Microsoft gave FBI a set of BitLocker encryption keys to unlock suspects’ laptops: Reports

Microsoft gave FBI a set of BitLocker encryption keys to unlock suspects’ laptops: Reports

Capital One acquires Brex for a steep discount to its peak valuation, but early believers are laughing all the way to the bank

Capital One acquires Brex for a steep discount to its peak valuation, but early believers are laughing all the way to the bank

Not to be outdone by OpenAI, Apple is reportedly developing an AI wearable

Not to be outdone by OpenAI, Apple is reportedly developing an AI wearable

Anthropic’s CEO stuns Davos with Nvidia criticism

Anthropic’s CEO stuns Davos with Nvidia criticism

Image

© 2025 TechCrunch Media LLC.