(Yicai) July 28 -- Chinese startup Moonshot AI released the complete model weights for its new Kimi K3 artificial intelligence model, with its 2.8 trillion parameters making it the world's first open-source model in the 3- trillion-parameter class.

Beijing-based Moonshot AI released the model weights on the global open-source community Hugging Face yesterday, along with a detailed technical report and three core training infrastructure technologies: MoonEP, FlashKDA, and AgentEnv. Kimi K3 is under a modified MIT license, allowing developers to download, modify, and deploy it locally.

In addition, Kimi K3's testing environment covered Nvidia H200 and general-purpose computing on graphics processing units from other vendors, indicating that its deployment capabilities are not limited to the ecosystem of the US chip giant.

Kimi K3 has 104 billion active parameters and is based on a Mixture of Experts architecture, where each token activates only 16 out of 896 routing experts. It employs a self-developed Kimi Delta Attention hybrid linear attention mechanism along with Attention Residuals tech, which natively supports visual understanding and has a context window of up to one million tokens.

Compared to the previous generation Kimi K2, the overall scaling efficiency has improved by about 2.5 times.

Kimi K3 outperformed all open-source and most closed-source models across dozens of benchmark tests in programming, agents, reasoning, vision, and other areas, according to the technical report. In Arena Intelligence's WebDev Arena, it achieved an Elo score of 1,679, becoming the first open-source model to claim the top spot on the leaderboard.

In addition, Kimi K3 scored 57.1 on the Artificial Analysis Intelligence Index v4.1, ranking fourth worldwide. However, its overall performance still slightly lags behind Claude Fable 5 and GPT-5.6 Sol, the report said.

The open source initiative aims to accelerate the deployment and popularization of cutting-edge intelligence, as well as to promote research in artificial general intelligence, Moonshot AI noted. With the simultaneous release of Kimi K3's complete weights, technical report, and core infrastructure technologies, open-source large language models have officially entered the 2.8-trillion-parameter stage, it stressed.

Editor: Martin Kadiev