Baidu Releases ERNIE 5.5 Flash Base: A 280B MoE Base Model With 14B Active Parameters
Covered by 2 sources
Read the full article
Baidu released ERNIE 5.5 Flash Base, a 280B-parameter sparse Mixture-of-Experts language model activating 14B parameters per token, designed as a raw pre-training checkpoint for fine-tuning and evaluation.
