openbmb/Ultra-FineWeb-L3
Viewer • Updated • 1.06B • 19k • 346
A 59M parameters model with Deepseek V4.1 Flash architechture
Other formats:
Trained on 1B tokens of Ultra-FineWeb-L3 over 1 day 12 hours using 1x RTX 5060 Ti costing us $5.64.
Official context window is 8192 tokens. However, the model's native context window is 1M tokens. Tokens above 8192 has never been trained.