You need to agree to share your contact information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

Log in or Sign Up to review the conditions and access this model content.

INCOMPLETE SAGE TP4 expert shards (not a loadable EXL3 pack)

Per-MoE-layer compiled expert shards (~6 GiB). Final model-XXXXX-of-YYYYY rename + native/Engram still pending. Flagship recipe SHA256 ad31c2e18596ea15cd77a4ae822aa00b5a380cfa70f07310dbc9ce9f50aa4762. Not identity-H. Public + gated=manual. Not a loadable EXL3 pack.

Downloads last month
6
Safetensors
Model size
325B params
Tensor type
F32
·
BF16
·
F8_E4M3
·
I8
·
F16
·
I16
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support