Baidu open-sources ERNIE-5.5-Flash-Base: 280B MoE, 33.5T tokens, 1M context
AdinaYakup · x · 2026-11-05
Baidu released ERNIE-5.5-Flash-Base, an open MoE base model with 280B total params (14B active), 33.5T pretraining tokens, up to 1M context, and hybrid sparse attention, shipped as a raw base checkpoint under a custom license.
Official benchmarks vs DeepSeek-V4-Flash-Base, Hy3-preview-Base and MiMo-V2.5-Base show it leading on most tasks: MMLU 93.3, CMMLU 90.0, BBH(CoT) 89.2, HumanEval+ 82.3, MBPP+ 81.5, plus long-context RepoQA 91.7 and InfiniteBench 46.5. Available on Hugging Face (incl. BF16 weights).
More from Models
- Users report models feeling smarter ahead of expected OpenAI release tonight — xiaohu · 2026-11-05
- ByteDance open-sources OpenViking: agent memory that cuts input tokens up to 91% — Promptmethus · 2026-11-05
- New weight-space benchmark ranks GPT-6 Astra first at 96.3 for writing neural net weights from scratch — BorisMPower · 2026-11-05
- Qwen3.5-0.8B 'weight-space programming' finetune trends on Hugging Face — RikiyaT · 2026-11-05
- Hugging Face launches ML Intern, an open-source agent that does ML for you — Gradio · 2026-11-05
- ChatGPT, Claude and Gemini all meter usage now — none tells you what one message costs — Ok_Negotiation_2587 · 2026-11-05