Optimizing Qwen3 CPU ONLY inference on Tanzu Platform: Cloud Foundry Weekly: Ep 55

21/05/2025 1h 5min Episodio 55
Optimizing Qwen3 CPU ONLY inference on Tanzu Platform: Cloud Foundry Weekly: Ep 55

Listen "Optimizing Qwen3 CPU ONLY inference on Tanzu Platform: Cloud Foundry Weekly: Ep 55"

Episode Synopsis


Hot off the presses in model releases - we will explore the Qwen3-30b-a3b MoE model running on the Tanzu Platform. Early testing shows it performs exceptionally well on somewhat older enterprise-grade server CPUs (aka Cascade Lake). This show will provide some insights on how enterprises can use their existing server infrastructure to start their intelligent application modernization efforts.

More episodes of the podcast Cloud Foundry Weekly