glm4v_text¶


Model type	`glm4v_text`
Class	`Glm4CausalLMModel`
Task	`text-generation`
Source	`models/glm.py`

Description¶

GLM4 text model (4-norm decoder + fused gate_up_proj).

Uses Glm4DecoderLayer with pre+post norms on both sub-layers. gate_up_proj is kept fused — activations are split in the MLP forward pass (via :class:FusedGateUpMLP in Glm4DecoderLayer).

Usage¶

mobius build --model <MODEL_ID> output_dir/

from mobius import build

model = build("<MODEL_ID>")