百度开源 2800 亿参数 ERNIE 5.5 Flash Base 基础模型
Baidu releases ERNIE 5.5 Flash Base with 280 billion parameters as an open-source foundation model
百度发布开源基础模型 ERNIE-5.5-Flash-Base，刻意不做微调和后训练，由开发者自行决定后续方向，后训练版本后续推出。该模型采用 MoE 架构，总参数 2800 亿、每次输入激活 140 亿，预训练使用 33.5 万亿 token，训练上下文最高 128000 token，设计上支持微调与推理时最高 100 万 token 上下文。百度称其在推理、代码和长上下文基准上与 DeepSeek V4 Flash、MiMo V2.5 等同规模开源基础模型竞争，权重已在 Hugging Face 以 FP8 和 BF16 提供。
正文
Baidu's AI development is still alive, but most of the work is on you. The Chinese company has released ERNIE-5.5-Flash-Base , an open foundation model that ships deliberately without fine-tuning or post-training. Developers are expected to decide for themselves where to take the model. A post-trained version is supposed to follow.
ERNIE 5.5 uses a mixture-of-experts architecture with 280 billion parameters, 14 billion of which are active per input. The model was pre-trained on 33.5 trillion tokens with a training context of up to 128,000 tokens, though it's designed to support a context of up to one million tokens during fine-tuning and inference.
In benchmarks for reasoning, code, and long-context tasks, the model competes with similarly sized open foundation models like DeepSeek V4 Flash or MiMo V2.5, according to Baidu. The weights are available on Hugging Face in FP8 and BF16.