{"version":1,"type":"story","url":"https://digestai.news/story/baidu-releases-ernie-5-5-flash-base-for-training-with-280b-parameters","json":"https://digestai.news/story/baidu-releases-ernie-5-5-flash-base-for-training-with-280b-parameters.json","markdown":"https://digestai.news/story/baidu-releases-ernie-5-5-flash-base-for-training-with-280b-parameters.md","slug":"baidu-releases-ernie-5-5-flash-base-for-training-with-280b-parameters","headline":"Baidu releases ERNIE 5.5 Flash Base for training with 280B parameters","summary":"Baidu’s ERNIE team released ERNIE 5.5 Flash Base on November 5, 2026, offering downloadable weights for a text-only mixture-of-experts model. The model has 280 billion total parameters and 14 billion activated per token, ending pre-training to serve as a starting point for domain adaptation, instruction tuning, and reinforcement-learning research. It supports 1 million tokens of context and includes a 43-layer architecture with sparse attention mechanisms to reduce computational costs as context grows.\n\nThe release also provides a fixed baseline for comparing training methods. Baidu claims speed gains of up to 175% in forward passes and 74% in backward passes over FlashAttention-4 on NVIDIA’s SM100 hardware. Benchmarks against models like DeepSeek-V4-Flash-Base show mixed results, with ERNIE leading in MMLU and HumanEval+ but trailing in FullStackBench and InfiniteBench. The model is licensed under territorial and commercial restrictions, excluding the EU, UK, and South Korea, and requiring prior authorization for commercial use beyond $50 million in annual revenue.","keyPoints":["ERNIE 5.5 Flash Base is a 280B-parameter text-only MoE model with 1M-token context support","Baidu claims 175% speed gains in forward passes on NVIDIA’s SM100 hardware","License excludes EU, UK, South Korea and restricts commercial use beyond $50M revenue"],"whyItMatters":"This release offers researchers a standardized checkpoint for long-context, efficient training, but its hardware demands and licensing limits practical adoption outside Baidu’s supported regions and revenue thresholds.","category":{"slug":"research","name":"Research","url":"https://digestai.news/category/research"},"entities":{"companies":["Baidu","NVIDIA"],"models":["ERNIE 5.5 Flash Base","DeepSeek-V4-Flash-Base","Hy3-preview-Base","MiMo-V2.5-Base"],"people":[]},"firstPublishedAt":"2026-11-05T10:25:04Z","updatedAt":"2026-11-05T10:25:04Z","sourceCount":1,"hasPrimarySource":false,"sources":[{"outlet":"Unite.AI","title":"Baidu Releases ERNIE 5.5 Flash Base for Further Training","url":"https://unite.ai/baidu-releases-ernie-5-5-flash-base-for-further-training","publishedAt":"2026-11-05T10:25:04Z","type":"press","primary":false,"lead":true}],"sourceNotes":null,"discussions":[],"thread":null,"cite":{"text":"Digest AI, \"Baidu releases ERNIE 5.5 Flash Base for training with 280B parameters\", 5 November 2026, https://digestai.news/story/baidu-releases-ernie-5-5-flash-base-for-training-with-280b-parameters","publisher":"Digest AI","title":"Baidu releases ERNIE 5.5 Flash Base for training with 280B parameters","datePublished":"2026-11-05T10:25:04Z","url":"https://digestai.news/story/baidu-releases-ernie-5-5-flash-base-for-training-with-280b-parameters"},"generatedBy":"Written by Digest AI's editorial model from the linked sources; the sources are the record.","license":"Headlines, digests and key points are written by Digest AI and may be quoted with a link to the story page. Linked articles belong to their publishers. Terms: https://digestai.news/terms#reuse"}