hunyuan_v1_dense¶
Model type |
|
Class |
|
Task |
|
Source |
|
Description¶
HunYuan V1 Dense text model.
HF HunYuanV1Dense unconditionally applies per-head RMSNorm on Q/K
projections (query_layernorm / key_layernorm). We enable
attn_qk_norm=True in the config and rename the HF weight keys.
Usage¶
mobius build --model <MODEL_ID> output_dir/
from mobius import build
model = build("<MODEL_ID>")