#MoE

OpenAI released a privacy filter model (openai/privacy-filter) using a Mixture of Experts (MoE) architecture with 50M active parameters and 1.5B total parameters. Designed to filter private information from trillion-scal…

Qwen3.6-35B-A3B is a sparse Mixture-of-Experts (MoE) language model with 35B total parameters and 3B active parameters per token during inference. Released under Apache 2.0 license. Supports agentic coding workflows with…