LaunchAI ModelsNovember 5, 2026

Baidu releases ERNIE 5.5 Flash Base, a 280B-parameter MoE

Read original source →ernie.baidu.com

Sparse MoE with 280B total parameters, 14B activated per token, pre-trained from scratch on 33.5T tokens with up to 128K training context. Weights are the end-of-pre-training checkpoint, released for domain adaptation, instruction tuning and RL research, with 1M-token context supported for fine-tuning and inference.

2 sources

More stories today

Open the live feed