NVIDIA releases Dory1.0-4B, a looped 4B reasoning model
The BF16 checkpoint uses a 256K context, OpenMDW-1.1, and variable loop counts without retraining.
1 stories, newest first · one card per story
The BF16 checkpoint uses a 256K context, OpenMDW-1.1, and variable loop counts without retraining.
Today 2.1K readers · 2.6K views · 393 reads · all time 108K views