You signed in with another tab or window.Reload to refresh your session.You signed out in another tab or window.Reload to refresh your session.You switched accounts on another tab or window.Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: CHANGELOG.rst
+9-18Lines changed: 9 additions & 18 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -1,24 +1,7 @@
1
1
Model Optimizer Changelog (Linux)
2
2
=================================
3
-
0.41 (2025-12-xx)
4
-
^^^^^^^^^^^^^^^^^
5
-
6
-
**Deprecations**
7
-
8
-
**New Features**
9
-
10
-
- Add support for PyTorch Geometric quantization.
11
-
12
-
**Documentation**
13
-
14
-
- Deprecate ``examples/megatron-lm`` in favor of more detailed documentation in `Megatron-LM/examples/post_training/modelopt<https://github.com/NVIDIA/Megatron-LM/tree/main/examples/post_training/modelopt>`_.
15
-
16
-
**Misc**
17
-
18
-
- Bump minimum recommended transformers version to 4.53.
19
3
20
-
21
-
0.40 (2025-12-xx)
4
+
0.40 (2025-12-11)
22
5
^^^^^^^^^^^^^^^^^
23
6
24
7
**Bug Fixes**
@@ -32,7 +15,15 @@ Model Optimizer Changelog (Linux)
32
15
- Add ``specdec_bench`` example to benchmark speculative decoding performance. See `examples/specdec_bench/README.md<https://github.com/NVIDIA/TensorRT-Model-Optimizer/tree/main/examples/specdec_bench#speculative-decoding-benchmark>`_ for more details.
33
16
- Add FP8/NVFP4 KV cache quantization support for Megatron Core models.
34
17
- Add flag ``trt_plugins_precision`` in ONNX autocast to indicate custom ops precision. This is similar to the flag already existing in the quantization workflow.
18
+
- Add support for PyTorch Geometric quantization.
19
+
20
+
**Documentation**
21
+
22
+
- Deprecate ``examples/megatron-lm`` in favor of more detailed documentation in `Megatron-LM/examples/post_training/modelopt<https://github.com/NVIDIA/Megatron-LM/tree/main/examples/post_training/modelopt>`_.
35
23
24
+
**Misc**
25
+
26
+
- Bump minimum recommended transformers version to 4.53.