fix export

Signed-off-by: Jennifer Chen <[email protected]>

Author: jenchen13

examples/nemo_run/qat/README.md

@@ -40,7 +40,7 @@ graph TD;
 
 ## Results
 
-QAT of Qwen3-8B NVFP4 recovers most of the accuracy on the MMLU benchmark after NVFP4 PTQ. We finetune the Qwen3-8B NVFP4 checkpoint for 200 steps with a learning rate of 1e-5 and global batch size of 512.
+QAT of Qwen3-8B NVFP4 recovers most of the accuracy on the MMLU benchmark after NVFP4 PTQ. We finetune the Qwen3-8B NVFP4 checkpoint for 200 steps with a learning rate of 1e-5 and global batch size of 512 on one node of 8 x H100 GPUs.
 
 |                           | MMLU 5% |
 |---------------------------|---------|

modelopt/torch/export/plugins/nemo_run.py

@@ -24,14 +24,12 @@
 def export_most_recent_ckpt(directory: str, output_path: str):
     """Export most recent checkpoint from a NeMo Run experiment directory."""
     most_recent_ckpt = _get_most_recent_ckpt(directory)
-    modelopt_kwargs = {"export_extra_modules": True}
     logging.info(f"Exporting most recent NeMo Run checkpoint: {most_recent_ckpt}")
     export_ckpt(
         most_recent_ckpt,
         "hf",
         output_path=output_path,
         overwrite=True,
-        modelopt_export_kwargs=modelopt_kwargs,
     )