Skip to content
Inteligencia IASep 18, 2026Inteligencia IA
Articulo

Zhipu AI launched GLM-5.3-FlashX, claiming nearly 200 tokens per second inference on ~100,000 domestic accelerators.

The base Flash model is a 320B MoE with 18B active parameters, optimized by an Infra Agent serving stack.

Generado por IA: resúmenes redactados por IA a partir de las fuentes enlazadas. Cómo usamos la IA

Generado por IAFuente: Pandaily
01

Resumen fuente

Zhipu AI launched GLM-5.3-FlashX, claiming nearly 200 tokens per second inference on ~100,000 domestic accelerators. The base Flash model is a 320B MoE with 18B active parameters, optimized by an Infra Agent serving stack.