Baidu Releases ERNIE-5.5-Flash-Base Foundation Model with 280B Total Parameters and 14B Activated

The Baidu ERNIE team has released ERNIE-5.5-Flash-Base, a sparse MoE language model with 280B total parameters and 14B activated per token, pretrained from scratch on 33.5T tokens, with a maximum training context of 128K and support for 1M-token context in fine-tuning and inference. The released weights are the checkpoint at the end of pretraining, without mid-training or post-training, and are available on Hugging Face in FP8 and BF16 safetensors formats for domain adaptation, instruction fine-tuning, and reinforcement learning research. The official comparison with DeepSeek-V4-Flash-Base, Hy3-preview-Base, and MiMo-V2.5-Base uses the same evaluation framework: MMLU 93.3, MMLU-Pro 78.4, HumanEval+ 82.3, and LiveCodeBench 40.1.