Add GLM-4.7-Flash Lite MoE calibration wrapper

GLM-4.7-Flash Lite stores routed experts as packed 3D tensors in
Glm4MoeLiteNaiveMoe, so the existing calibration path not only skipped
MoE-aware calibration but also kept routed experts invisible to
Linear-targeted quantization.

Unpack the routed experts into per-expert Glm4MoeLiteMLP modules,
preserve the unpacked structure for quantization and checkpoint save,
and add focused modeling tests for expert activation, output parity,
and Linear visibility.

Signed-off-by: Jason Lu <Nottlespike@users.noreply.github.com>

Author: Nottlespike

src/llmcompressor/modeling/__init__.py

@@ -12,6 +12,7 @@
 # trigger registration
 from .deepseek_v3 import CalibrationDeepseekV3MoE  # noqa: F401
 from .glm4_moe import CalibrationGlm4MoeMoE  # noqa: F401
+from .glm4_moe_lite import CalibrationGlm4MoeLiteMoE  # noqa: F401
 from .glm_moe_dsa import CalibrationGlmMoeDsaMoE  # noqa: F401
 from .llama4 import SequentialLlama4TextMoe  # noqa: F401
 from .qwen3_moe import CalibrationQwen3MoeSparseMoeBlock  # noqa: F401

src/llmcompressor/modeling/glm4_moe_lite.py

@@ -0,0 +1,169 @@
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+
+import torch
+
+from llmcompressor.modeling.moe_context import MoECalibrationModule
+from llmcompressor.utils.dev import skip_weights_initialize
+
+if TYPE_CHECKING:
+    from transformers.models.glm4_moe_lite.configuration_glm4_moe_lite import (
+        Glm4MoeLiteConfig,
+    )
+    from transformers.models.glm4_moe_lite.modeling_glm4_moe_lite import (
+        Glm4MoeLiteMoE,
+        Glm4MoeLiteNaiveMoe,
+    )
+
+
+@MoECalibrationModule.register("Glm4MoeLiteMoE")
+class CalibrationGlm4MoeLiteMoE(MoECalibrationModule):
+    """
+    Calibration version of Glm4MoeLiteMoE that unfuses 3D expert parameters into
+    individual MLP modules (nn.Linear) so they can be quantized.
+
+    GLM-4.7-Flash Lite stores routed experts in a `Glm4MoeLiteNaiveMoe` module
+    using 3D parameters (`gate_up_proj`, `down_proj`) instead of `nn.Linear`
+    submodules. Since llm-compressor targets `Linear` modules, the original routed
+    experts are invisible to quantization and remain BF16 unless they are unpacked.
+
+    is_permanent = True so the unpacked `nn.Linear` expert structure persists for
+    quantization and checkpoint save.
+    """
+
+    is_permanent = True
+
+    def __init__(
+        self,
+        original: Glm4MoeLiteMoE,
+        config: Glm4MoeLiteConfig,
+        calibrate_all_experts: bool = True,
+        num_calibrate_experts: int | None = None,
+    ):
+        super().__init__()
+        self.top_k = config.num_experts_per_tok
+        self.num_experts = config.n_routed_experts
+        self.n_routed_experts = config.n_routed_experts
+        self.n_group = config.n_group
+        self.topk_group = config.topk_group
+        self.norm_topk_prob = config.norm_topk_prob
+        self.routed_scaling_factor = config.routed_scaling_factor
+
+        self.experts = SequentialGlm4MoeLiteExperts(config, original.experts)
+        self.gate = original.gate
+        self.shared_experts = original.shared_experts
+        self.calibrate_all_experts = calibrate_all_experts
+
+    def route_tokens_to_experts(
+        self, router_logits: torch.Tensor
+    ) -> tuple[torch.Tensor, torch.Tensor]:
+        """Match the Hugging Face GLM-4.7-Flash Lite group router."""
+        router_logits = router_logits.sigmoid()
+        router_logits_for_choice = router_logits + self.gate.e_score_correction_bias
+        group_scores = (
+            router_logits_for_choice.view(
+                -1, self.n_group, self.n_routed_experts // self.n_group
+            )
+            .topk(2, dim=-1)[0]
+            .sum(dim=-1)
+        )
+        group_idx = torch.topk(group_scores, k=self.topk_group, dim=-1, sorted=False)[1]
+        group_mask = torch.zeros_like(group_scores)
+        group_mask.scatter_(1, group_idx, 1)
+        score_mask = (
+            group_mask.unsqueeze(-1)
+            .expand(-1, self.n_group, self.n_routed_experts // self.n_group)
+            .reshape(-1, self.n_routed_experts)
+        )
+        scores_for_choice = router_logits_for_choice.masked_fill(
+            ~score_mask.bool(), 0.0
+        )
+        topk_indices = torch.topk(
+            scores_for_choice, k=self.top_k, dim=-1, sorted=False
+        )[1]
+        topk_weights = router_logits.gather(1, topk_indices)
+        if self.norm_topk_prob:
+            denominator = topk_weights.sum(dim=-1, keepdim=True) + 1e-20
+            topk_weights /= denominator
+        topk_weights = topk_weights * self.routed_scaling_factor
+        return topk_indices, topk_weights
+
+    def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
+        residuals = hidden_states
+        orig_shape = hidden_states.shape
+        router_logits = self.gate(hidden_states)
+        topk_indices, topk_weights = self.route_tokens_to_experts(router_logits)
+        hidden_states = hidden_states.view(-1, hidden_states.shape[-1])
+
+        # Run unpacked experts sequentially so routed MLPs stay visible
+        # as Linear modules.
+        final_hidden_states = torch.zeros_like(hidden_states, dtype=topk_weights.dtype)
+        with torch.no_grad():
+            expert_mask = torch.nn.functional.one_hot(
+                topk_indices, num_classes=self.num_experts
+            )
+            expert_mask = expert_mask.permute(2, 1, 0)
+
+        for i in range(self.num_experts):
+            top_k_pos, token_idx = torch.where(expert_mask[i])
+            has_tokens = token_idx.numel() > 0
+
+            if self.calibrate_all_experts:
+                expert_out_all = self.experts[i](hidden_states)
+                if not has_tokens:
+                    continue
+                expert_out = expert_out_all[token_idx]
+            else:
+                if not has_tokens:
+                    continue
+                expert_out = self.experts[i](hidden_states[token_idx])
+
+            weighted_output = expert_out * topk_weights[token_idx, top_k_pos, None]
+            final_hidden_states.index_add_(
+                0, token_idx, weighted_output.to(final_hidden_states.dtype)
+            )
+
+        hidden_states = final_hidden_states.type(hidden_states.dtype).view(*orig_shape)
+        hidden_states = hidden_states + self.shared_experts(residuals)
+        return hidden_states
+
+    def restore(self, original: torch.nn.Module) -> torch.nn.Module:
+        """Keep the unpacked expert structure for quantization and checkpoint save."""
+        return self
+
+
+class SequentialGlm4MoeLiteExperts(torch.nn.ModuleList):
+    """
+    Unpacks 3D expert parameter tensors into individual Glm4MoeLiteMLP modules so
+    each routed expert has standard `nn.Linear` projections visible to
+    `targets="Linear"`.
+    """
+
+    def __init__(self, config: Glm4MoeLiteConfig, original: Glm4MoeLiteNaiveMoe):
+        from transformers.models.glm4_moe_lite.modeling_glm4_moe_lite import (
+            Glm4MoeLiteMLP,
+        )
+
+        self.num_experts = config.n_routed_experts
+        intermediate_size = config.moe_intermediate_size
+
+        with skip_weights_initialize():
+            super().__init__(
+                [
+                    Glm4MoeLiteMLP(config, intermediate_size=intermediate_size)
+                    for _ in range(self.num_experts)
+                ]
+            )
+
+        gate_up_data = original.gate_up_proj.data
+        down_data = original.down_proj.data
+
+        for i in range(self.num_experts):
+            gate_up = gate_up_data[i]
+            down = down_data[i]
+            gate_proj, up_proj = gate_up.chunk(2, dim=0)
+
+            self[i].gate_proj.weight.data = gate_proj.contiguous()
+            self[i].up_proj.weight.data = up_proj.contiguous()
+            self[i].down_proj.weight.data = down.contiguous()

tests/llmcompressor/modeling/test_calib_glm4_moe_lite.py

@@ -0,0 +1,115 @@
+from functools import partial
+
+import pytest
+import torch
+
+_glm_cfg = pytest.importorskip(
+    "transformers.models.glm4_moe_lite.configuration_glm4_moe_lite",
+    reason="glm4_moe_lite requires transformers >= 5.x",
+)
+_glm_mod = pytest.importorskip(
+    "transformers.models.glm4_moe_lite.modeling_glm4_moe_lite",
+    reason="glm4_moe_lite requires transformers >= 5.x",
+)
+Glm4MoeLiteConfig = _glm_cfg.Glm4MoeLiteConfig
+Glm4MoeLiteMoE = _glm_mod.Glm4MoeLiteMoE
+
+from llmcompressor.modeling.glm4_moe_lite import CalibrationGlm4MoeLiteMoE  # noqa: E402
+from llmcompressor.utils.helpers import calibration_forward_context  # noqa: E402
+from tests.testing_utils import requires_gpu  # noqa: E402
+
+
+def _tiny_config():
+    """Small config for fast unit tests (8 experts instead of 64)."""
+    return Glm4MoeLiteConfig(
+        hidden_size=64,
+        intermediate_size=128,
+        moe_intermediate_size=32,
+        n_routed_experts=8,
+        num_experts_per_tok=2,
+        n_shared_experts=1,
+        n_group=1,
+        topk_group=1,
+        num_hidden_layers=4,
+        num_attention_heads=4,
+        num_key_value_heads=4,
+        vocab_size=256,
+        pad_token_id=0,
+        eos_token_id=[0],
+    )
+
+
+@requires_gpu
+def test_calib_glm4_moe_lite_all_experts_triggered():
+    config = _tiny_config()
+    with torch.device("cuda"):
+        original = Glm4MoeLiteMoE(config)
+        for param in original.parameters():
+            param.data.normal_(mean=0.0, std=0.02)
+
+    module = CalibrationGlm4MoeLiteMoE(original, config, calibrate_all_experts=True)
+
+    num_experts = len(module.experts)
+    expert_triggered = [False for _ in range(num_experts)]
+
+    def hook_fn(i, module, input, output):
+        expert_triggered[i] = True
+
+    for i, expert in enumerate(module.experts):
+        expert.register_forward_hook(partial(hook_fn, i))
+
+    hidden_dim = config.hidden_size
+    batch, seq_len = 4, 32
+    sample = torch.randn(batch, seq_len, hidden_dim, device="cuda")
+
+    with calibration_forward_context(module):
+        with torch.no_grad():
+            _ = module(sample)
+
+    assert all(expert_triggered), f"Not all experts were triggered: {expert_triggered}"
+
+
+@requires_gpu
+def test_calib_glm4_moe_lite_output_matches():
+    config = _tiny_config()
+    with torch.device("cuda"):
+        original = Glm4MoeLiteMoE(config)
+        for param in original.parameters():
+            param.data.normal_(mean=0.0, std=0.02)
+
+    hidden_dim = config.hidden_size
+    batch, seq_len = 4, 32
+    sample = torch.randn(batch, seq_len, hidden_dim, device="cuda")
+
+    with calibration_forward_context(original):
+        true_out = original(sample)
+
+    module = CalibrationGlm4MoeLiteMoE(original, config, calibrate_all_experts=True)
+    with calibration_forward_context(module):
+        out = module(sample)
+        assert torch.nn.functional.mse_loss(true_out, out) < 0.1
+
+    module = CalibrationGlm4MoeLiteMoE(original, config, calibrate_all_experts=False)
+    with calibration_forward_context(module):
+        out = module(sample)
+        assert torch.nn.functional.mse_loss(true_out, out) < 0.1
+
+
+@requires_gpu
+def test_calib_glm4_moe_lite_experts_are_linear():
+    """Verify that unpacked routed experts expose Linear modules for quantization."""
+    config = _tiny_config()
+    with torch.device("cuda"):
+        original = Glm4MoeLiteMoE(config)
+
+    module = CalibrationGlm4MoeLiteMoE(original, config, calibrate_all_experts=True)
+
+    linear_names = [
+        name for name, mod in module.named_modules() if isinstance(mod, torch.nn.Linear)
+    ]
+    expected_expert_linears = config.n_routed_experts * 3
+    expected_shared_linears = 3
+    assert len(linear_names) == expected_expert_linears + expected_shared_linears, (
+        f"Expected {expected_expert_linears + expected_shared_linears} Linear modules, "
+        f"found {len(linear_names)}: {linear_names}"
+    )