Skip to content

[Model] Modify the gate implementation of glm4_moe #22832

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 2 commits into from
Aug 14, 2025
Merged

[Model] Modify the gate implementation of glm4_moe #22832

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
bbeb4e7
Done
jeejeelee Aug 13, 2025
1800962
Merge branch 'vllm-project:main' into fix-glm4-gate
jeejeelee Aug 14, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion docs/models/supported_models.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -615,7 +615,7 @@ These models primarily accept the [`LLM.generate`](./generative_models.md#llmgen
| `Gemma3nForConditionalGeneration` | Gemma 3n | T + I + A | `google/gemma-3n-E2B-it`, `google/gemma-3n-E4B-it`, etc. | | | ✅︎ |
| `GLM4VForCausalLM`<sup>^</sup> | GLM-4V | T + I | `zai-org/glm-4v-9b`, `zai-org/cogagent-9b-20241220`, etc. | ✅︎ | ✅︎ | ✅︎ |
| `Glm4vForConditionalGeneration` | GLM-4.1V-Thinking | T + I<sup>E+</sup> + V<sup>E+</sup> | `zai-org/GLM-4.1V-9B-Thinking`, etc. | ✅︎ | ✅︎ | ✅︎ |
| `Glm4vMoeForConditionalGeneration` | GLM-4.5V | T + I<sup>E+</sup> + V<sup>E+</sup> | `zai-org/GLM-4.5V`, etc. | | ✅︎ | ✅︎ |
| `Glm4vMoeForConditionalGeneration` | GLM-4.5V | T + I<sup>E+</sup> + V<sup>E+</sup> | `zai-org/GLM-4.5V`, etc. | ✅︎ | ✅︎ | ✅︎ |
| `GraniteSpeechForConditionalGeneration` | Granite Speech | T + A | `ibm-granite/granite-speech-3.3-8b` | ✅︎ | ✅︎ | ✅︎ |
| `H2OVLChatModel` | H2OVL | T + I<sup>E+</sup> | `h2oai/h2ovl-mississippi-800m`, `h2oai/h2ovl-mississippi-2b`, etc. | | ✅︎ | ✅︎ |
| `Idefics3ForConditionalGeneration` | Idefics3 | T + I | `HuggingFaceM4/Idefics3-8B-Llama3`, etc. | ✅︎ | | ✅︎ |
Expand Down
20 changes: 10 additions & 10 deletions vllm/model_executor/models/glm4_moe.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -41,7 +41,6 @@
from vllm.model_executor.layers.layernorm import RMSNorm
from vllm.model_executor.layers.linear import (MergedColumnParallelLinear,
QKVParallelLinear,
ReplicatedLinear,
RowParallelLinear)
from vllm.model_executor.layers.logits_processor import LogitsProcessor
from vllm.model_executor.layers.quantization import QuantizationConfig
Expand Down Expand Up @@ -118,14 +117,15 @@ def __init__(
if config.hidden_act != "silu":
raise ValueError(f"Unsupported activation: {config.hidden_act}. "
"Only silu is supported for now.")

self.gate = ReplicatedLinear(config.hidden_size,
config.n_routed_experts,
bias=False,
quant_config=None,
params_dtype=torch.float32,
prefix=f"{prefix}.gate")

# NOTE In the transformers implementation, the gate isn't an nn.Linear,
# so we cannot use ReplicatedLinear here.
# See: https://github.com/huggingface/transformers/blob/v4.55.1/src/transformers/models/glm4_moe/modeling_glm4_moe.py#L260
self.gate = nn.Linear(
config.hidden_size,
config.n_routed_experts,
bias=False,
dtype=torch.float32,
)
self.gate.e_score_correction_bias = nn.Parameter(
torch.empty(config.n_routed_experts, dtype=torch.float32))

Expand Down Expand Up @@ -181,7 +181,7 @@ def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:

if self.n_shared_experts is not None:
shared_output = self.shared_experts(hidden_states)
router_logits, _ = self.gate(hidden_states.to(dtype=torch.float32))
router_logits = self.gate(hidden_states.to(dtype=torch.float32))
final_hidden_states = self.experts(
hidden_states=hidden_states,
router_logits=router_logits) * self.routed_scaling_factor
Expand Down