Baidu Open-Sources ERNIE 5.5 Flash Base, Setting SOTA for Same-Size Models on Multiple Benchmarks
English text is being prepared.
On November 5, Baidu officially open-sourced the base model ERNIE 5.5 Flash Base, which adopts a MoE architecture with a total parameter count of 280 billion and 14 billion parameters activated per token. The model was pre-trained on 33.5 trillion tokens, with a maximum pre-training context length of 128K, and supports a context length of up to 1 million tokens during fine-tuning and inference. The company claims it achieves SOTA among same-size models on benchmarks covering knowledge, reasoning, mathematics, code, multilingual capabilities, and long context, surpassing models such as DeepSeek-V4-Flash-Base, Hy3-preview-Base, and MiMo-V2.5-Base.
Why read this
AI score 73
The original text provides the parameter count, activation scale, and pre-training data volume of ERNIE 5.5 Flash Base, allowing comparison of specification differences with same-size open-source base models.
Topics
Body
11月5日，百度正式开源最新基座模型文心5.5 Flash Base（ERNIE 5.5 Flash Base）。模型采用MoE架构，总参数量2800亿，每个 Token 激活140亿参数，使用33.5万亿个 Token 完成预训练，预训练上下文长度最高达128K，微调与推理阶段支持最长100万 Token上下文。在知识、推理、数学、代码、多语言和长上下文等多个基准测试中取得同尺寸模型 SOTA 表现，超过 DeepSeek-V4-Flash-Base、Hy3-preview-Base、MiMo-V2.5-Base 等模型。
本文由百度提供，量子位获授权转载，观点归原作者所有。