Baidu Releases ERNIE 5.5 Flash Preview with 280B Parameters and Open-Weight Pledge
Baidu has released ERNIE 5.5 Flash Preview, the first post-trained variant of its Base model, featuring 280B total parameters, 14B activated parameters and a 1M-token context window, targeting coding, 3D modeling and content creation. The API is live on the Qianfan platform, with weights promised to be open-sourced by November 25; Baidu describes it as an intermediate post-training checkpoint still under rapid iteration.
Related Entities
Full Report
IT之家 November 11 news, Baidu's ERNIE large model today released ERNIE 5.5 Flash Preview, which is the first post-training version based on the previously released Base model.
The model has 280B total parameters, 14B activated parameters, and supports a 1M context. The official statement says it performs excellently in programming, 3D modeling, knowledge work, content creation, and other areas.
According to the introduction, ERNIE 5.5 Flash Preview balances intelligence level and inference efficiency. In multiple benchmark tests, ERNIE 5.5 Flash Preview achieved competitive scores while maintaining extremely fast decoding speed. The model API is now available on the Qianfan Platform, and the weights will be open-sourced before November 25.
IT之家 attaches the benchmark test results as follows:
The official statement says that ERNIE 5.5 Flash Preview is an intermediate version in the post-training process. Although it has achieved relatively competitive performance among models of the same size, it may still have some capability deficiencies caused by insufficient training. Post-training is still ongoing, and ERNIE 5.5 Flash is still evolving rapidly. The team will closely monitor the actual performance and user feedback of the Preview model in real-world scenarios, and will make optimizations in subsequent official versions.