What Baidu released
Baidu's ERNIE site, under "Product updates / November 5, 2026", states: "Today we are releasing ERNIE-5.5-Flash-Base, a sparse Mixture-of-Experts (MoE) language model with 280B total parameters, 14B of which are activated per token." It was "pre-trained from scratch on 33.5T tokens with a training context of up to 128K, and supports a 1M-token context for fine-tuning and inference."
"The weights are the checkpoint at the end of pre-training, before any mid- or post-training" and "can serve as a common starting point for domain adaptation, instruction tuning and reinforcement-learning research." Baidu adds: "The post-trained ERNIE 5.5 Flash will follow."
Where to download it
Baidu writes that ERNIE-5.5-Flash-Base "is available on Hugging Face in FP8 and BF16 safetensors (the BF16 version is ERNIE-5.5-Flash-Base-BF16), under the ERNIE 5.5 Flash Base Model License Agreement v1.0." The Hugging Face model card for baidu/ERNIE-5.5-Flash-Base confirms the same sizes, context lengths and license name. Read that license before any commercial use.
Baidu's own benchmarks
Baidu's evaluation section compares the base checkpoint with DeepSeek-V4-Flash-Base, Hy3-preview-Base and MiMo-V2.5-Base under one harness. Selected numbers from Baidu's table for ERNIE-5.5-Flash-Base include MMLU 93.3, MMLU-Pro 78.4, HumanEval+ 82.3 and LiveCodeBench 40.1. These are vendor results, not independent tests.
What it means for you
If you fine-tune or continue-train open MoE bases, this is a new Baidu checkpoint to try under its license. It is not a chat-ready assistant out of the box. There is no ERNIE product profile on AppFor.ai yet; treat this news as the weight release only.
Go to the source
Sources & further reading
Based on the official documentation below. Features and access can change; check the provider before making a decision.
Was this page useful?
Help us make our coverage more useful. This is feedback on the page, not a rating of the tool.
Go a little further