Update

Signed-off-by: Chenjie Luo <chenjiel@nvidia.com>

Author: cjluo-nv

CHANGELOG.rst

@@ -8,7 +8,7 @@ NVIDIA Model Optimizer Changelog (Linux)
 
 - User does not need to manually register MOE modules to cover experts calibration coverage in PTQ workflow.
 - ``hf_ptq.py`` now saves the quantization summary and moe expert token count table to the export directory.
-- Add ``--moe_calib_experts_ratio`` flag in ``hf_ptq.py`` to specify the ratio of experts to calibrate during forward pass to improve expert coverage during calibration. Default to 1/4 of all the experts.
+- Add ``--moe_calib_experts_ratio`` flag in ``hf_ptq.py`` to specify the ratio of experts to calibrate during forward pass to improve expert coverage during calibration. Default to all the experts.
 - Add sparse attention optimization for transformer models (``modelopt.torch.sparsity.attention_sparsity``). This reduces computational cost by skipping attention computation. Supports calibration for threshold selection on HuggingFace models. See `examples/llm_sparsity/attention_sparsity/README.md <https://github.com/NVIDIA/Model-Optimizer/tree/main/examples/llm_sparsity/attention_sparsity>`_ for usage.
 
 0.42 (2026-02-xx)

examples/llm_ptq/example_utils.py

@@ -201,7 +201,7 @@ def build_quant_cfg(
     model_type,
     quant_cfg_choices,
     kv_quant_cfg_choices,
-    moe_calib_experts_ratio,
+    moe_calib_experts_ratio: float | None = None,
 ) -> dict[str, Any]:
     quant_cfg = {}
     assert qformat in quant_cfg_choices, (
@@ -234,13 +234,18 @@ def build_quant_cfg(
         )
 
     if moe_calib_experts_ratio:
+        assert 0 < moe_calib_experts_ratio <= 1, "moe_calib_experts_ratio must be between 0 and 1"
         if isinstance(quant_cfg["algorithm"], str):
             quant_cfg["algorithm"] = {
                 "method": quant_cfg["algorithm"],
                 "moe_calib_experts_ratio": moe_calib_experts_ratio,
             }
-        else:
+        elif isinstance(quant_cfg["algorithm"], dict):
             quant_cfg["algorithm"]["moe_calib_experts_ratio"] = moe_calib_experts_ratio
+        else:
+            warnings.warn(
+                f"Quantization algorithm: {quant_cfg['algorithm']} does not support setting moe_calib_experts_ratio"
+            )
 
     # Gemma 7B has accuracy regression using alpha 1. We set 0.5 instead.
     if model_type == "gemma" and "int8_sq" in qformat:

examples/llm_ptq/hf_ptq.py

@@ -1130,10 +1130,11 @@ def parse_args() -> argparse.Namespace:
     parser.add_argument(
         "--moe_calib_experts_ratio",
         type=float,
-        default=1.0 / 4,
+        default=1.0,
         help=(
             "Fraction of experts to calibrate during forward pass (ratio in (0.0, 1.0]). "
             "Only used for MOE models; used to reduce the number of experts calibrated during the forward pass."
+            "Does not impact non-MOE models."
         ),
     )
 

modelopt/torch/quantization/mode.py

@@ -227,6 +227,9 @@ def wrapped_calib_func(
 
     moe_calib_experts_ratio = kwargs.pop("moe_calib_experts_ratio", None)
     if moe_calib_experts_ratio is not None:
+        assert (
+            isinstance(moe_calib_experts_ratio, (int, float)) and 0 < moe_calib_experts_ratio <= 1
+        ), f"Invalid moe_calib_experts_ratio {moe_calib_experts_ratio!r}"
         for module in model.modules():
             if hasattr(module, "_moe_calib_experts_ratio"):
                 module._moe_calib_experts_ratio = moe_calib_experts_ratio

modelopt/torch/quantization/plugins/huggingface.py

@@ -499,7 +499,8 @@ def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
             assert 0 < self._moe_calib_experts_ratio <= 1, (
                 "moe_calib_experts_ratio must be between 0 and 1"
             )
-            # If any of the experts are in calibration mode, we will forward all tokens to all experts
+            # If any of the experts are in calibration mode, we will forward all tokens to
+            # self._moe_calib_experts_ratio % of the experts to improve the calibration coverage.
             # This is used only for calibration, we need to re-calculate the actual outputs again using
             # the original top_k
             if TRANSFORMERS_VERSION_GE_5_0: