k
keynews.ai
keynews.ai
FrontierThe story, in brief

Moonshot AI Releases Kimi K3: A 2.8 Trillion Parameter Open MoE Model With Kimi Delta Attention and 1M Context

2.8 trillion parameters. Moonshot AI just open-sourced Kimi K3 with 1M context and a novel attention mechanism that could reshape the MoE landscape.

Reporting by MarkTechPost 1 min brief
Illustration of a transparent lens revealing connected networks across layers of paper.
Exploring the next frontier of AI research.AI illustration by KeyNews
The KeyNews take

Why it matters

Moonshot AI's Kimi K3 represents a significant capability jump in open-source MoE architectures, combining massive scale (2.8T params), extended context (1M tokens), and novel attention innovations. This directly challenges closed-model incumbents and signals China's AI research leadership in efficiency-first design.

The key facts

7 to know

  1. 2.8 trillion parameters

  2. Mixture of Experts: 16 of 896 experts activated

  3. 1M token context window

  4. Novel architecture: Kimi Delta Attention + Attention Residuals

  5. Open-source release

  6. Published July 16, 2026

  7. Moonshot AI (Chinese lab)

Go to the source

MarkTechPostmarktechpost.com

Publisher excerpt: Moonshot AI released Kimi K3 on July 16, 2026. It is a 2.8-trillion-parameter open MoE model built on Kimi Delta Attention and Attention Residuals, activating 16 of 896 experts.
Read original report
Back to today's editionMore frontier news

The wider picture

View all
Illustration of a transparent lens revealing connected networks across layers of paper.
AI illustration by KeyNews
Frontier01

Alibaba Unveils Zhenwu V900 — and Plans Qwen Models With Up to 10 Trillion Parameters

Alibaba is advancing on two fronts simultaneously: announcing a custom AI accelerator (Zhenwu V900) and committing to massive model scale (10T parameters for future Qwen releases). For practitioners, this matters as a credible third-party capability play outside the US-China licensing squeeze; for enthusiasts, it's a significant lab-race signal about training compute and parameter scaling as competitive levers.

TechRepublic
Illustration of a transparent lens revealing connected networks across layers of paper.
AI illustration by KeyNews
Frontier02

Kyutai Releases Voice of Reason: A Speech-Native Model that Solves Spoken Math with Reinforcement Learning

A new capability frontier: models that reason directly in speech without transcription bottlenecks. This changes how we think about multimodal reasoning and what's possible with open-weight releases at scale.

MarkTechPost
Illustration of a transparent lens revealing connected networks across layers of paper.
AI illustration by KeyNews
Frontier03

Claude Opus 5.5, GPT-6 Sol, GPT-6 Luna, and a new price war

Frontier labs are shipping upgraded reasoning and multimodal models in rapid succession, signaling acceleration in the capability race. Simultaneous price cuts reshape AI economics for practitioners.

Simon Willison