Mixture-of-Experts

2026-09-07

Frontier Intelligence at Flash Cost: What GLM-5.3-Flash Actually Means

Z.ai just dropped GLM-5.3-Flash: 320B total, 18B active, MIT-licensed, natively multimodal, and it hit #1 of the week on OpenRouter before it even shipped. The interesting part isn't the benchmark — it's that the whole week of traffic ran on Chinese AI chips.

2026-09-03

Latent MoE: Compressing the Experts Before You Route Them

NVIDIA's Nemotron 3 Super compresses tokens into a low-rank latent space before routing to experts. Same inference cost, four times more specialists. That changes what MoE architectures are for.