OpenBMB Releases Ultra-FineWeb-L1, a 1T+ Token Web Corpus

OpenBMB has open-sourced Ultra-FineWeb-L1 on Hugging Face, a high-quality English web corpus of over 1 trillion tokens built from the latest Common Crawl snapshot, released under Apache 2.0 for LLM pretraining.

2026-08-20 ~ 2026-08-21 · 3 related posts

• OpenBMB Releases Ultra-FineWeb-L1, a Multi-Billion Token Web Corpus — openbmb · 2026-08-20
• OpenBMB releases Ultra-FineWeb-L1, a 1T+ token high-quality web dataset — zibuyu9 · 2026-08-21

1 near-duplicate retellings: AdinaYakup