VRAM is all we need
OpenMOSE
OpenMOSE
AI & ML interests
Can love be expressed as a tensor?
Organizations
None yet
VRAM is all we need
New hxa07D family of hybrid models, combining improved RWKV recurrent architectures with Transformer-based attention.
Designed for efficient long-cont
HXA079 family of hybrid models, combining RWKV recurrent architectures with Transformer-based attention.
Designed for efficient long-context.