Moonshot AI
(Beijing, China)
Builders of the Kimi family - flagship Kimi K3.1 Pro (3T-class, 1M context, always-on thinking; full open weights pledged by Nov 4, 2026) - widely used for agent orchestration and long-context production workloads.
What is the latest AI model from Moonshot AI?
Moonshot AI's primary flagship model is Kimi K3.1 Pro (3T-class MoE (post-trained from Kimi K3), 1,048,576 tokens context). First model of the K3.1 series and Moonshot's most capable: a 3T-class MoE post-trained from Kimi K3 with dramatically scaled joint RL across hundreds of domains, a 1M-token context, always-on thinking with max reasoning effort by default, image and video vision input, and automatic 5m/1h context caching at $2/$10 per 1M tokens; a kimi-k3.1-pro-highspeed variant ships 2-3x speed, and full open weights are pledged by Nov 4, 2026. How we define “latest”