update CHANGELOG

kaix-nv · kaix-nv · commit d2c05f2d4112 · 2025-10-10T15:57:11.000-07:00
Signed-off-by: Kai Xu &lt;kaix@nvidia.com&gt;
diff --git a/CHANGELOG.rst b/CHANGELOG.rst
@@ -1,6 +1,14 @@
 Model Optimizer Changelog (Linux)
 =================================
 
+0.41 (2025-12-xx)
+^^^^^^^^^^^^^^^^^
+
+**Deprecations**
+
+**New Features**
+- Add FP8/NVFP4 KV cache quantization support for Megatron Core models.
+
 0.39 (2025-11-xx)
 ^^^^^^^^^^^^^^^^^
 
diff --git a/modelopt/torch/quantization/plugins/__init__.py b/modelopt/torch/quantization/plugins/__init__.py
@@ -71,6 +71,3 @@
 
 with import_plugin("trl"):
     from .trl import *
-
-with import_plugin("mcore"):
-    from .mcore import *
