Tag

#mixture-of-experts

2 stories · “mixture-of-experts”.

An aerial view of an Amazon Web Services data center complex in the United States.

Sep 22, 2026

Moonshot's Kimi K3 Reaches Amazon Bedrock With Open-Weight Prompt Caching

Amazon Web Services added Moonshot AI's 2.8-trillion-parameter Kimi K3 to Bedrock on 18 September 2026, the first open-weight model there with prompt caching.

By TPW Editorial8 min read

Close-up of an NVIDIA DGX B200 HGX eight-way GPU baseboard, showing large stacked heatsinks and NVIDIA-labeled compute modules used to serve large AI models

Aug 6, 2026

Mixture of Experts, Explained: Why the Biggest AI Models Activate Only Part of Themselves

Sparse mixture-of-experts models activate a small fraction of their parameters per token. How routing, capacity and serving shape the largest AI models.

By TPW Editorial10 min read