nvidia/NVIDIA-Nemotron-Nano-9B-v2 · how to run on vllm with tool call parsing -- encountered an error

Unfortunately, when I add the commands for tool parsing and attempt to start vllm with vllm serve, I get:

    from vllm.entrypoints.cli.main import main
  File "/vllm/vllm/entrypoints/cli/__init__.py", line 4, in <module>
    from vllm.entrypoints.cli.benchmark.serve import BenchmarkServingSubcommand
  File "/vllm/vllm/entrypoints/cli/benchmark/serve.py", line 5, in <module>
    from vllm.benchmarks.serve import add_cli_args, main
  File "/vllm/vllm/benchmarks/serve.py", line 36, in <module>
    from vllm.benchmarks.datasets import (SampleRequest, add_dataset_parser,
  File "/vllm/vllm/benchmarks/datasets.py", line 32, in <module>
    from vllm.lora.utils import get_adapter_absolute_path
  File "/vllm/vllm/lora/utils.py", line 16, in <module>
    from vllm.lora.fully_sharded_layers import (
  File "/vllm/vllm/lora/fully_sharded_layers.py", line 15, in <module>
    from vllm.lora.layers import (ColumnParallelLinearWithLoRA,
  File "/vllm/vllm/lora/layers.py", line 23, in <module>
    from vllm.model_executor.layers.linear import (ColumnParallelLinear,
  File "/vllm/vllm/model_executor/layers/linear.py", line 22, in <module>
    from vllm.model_executor.layers.utils import dispatch_unquantized_gemm
  File "/vllm/vllm/model_executor/layers/utils.py", line 8, in <module>
    from vllm import _custom_ops as ops
  File "/vllm/vllm/_custom_ops.py", line 440, in <module>
    @register_fake("_C::marlin_qqq_gemm")
     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/torch/library.py", line 1023, in register
    use_lib._register_fake(op_name, func, _stacklevel=stacklevel + 1)
  File "/usr/local/lib/python3.12/site-packages/torch/library.py", line 214, in _register_fake
    handle = entry.fake_impl.register(func_to_register, source)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/torch/_library/fake_impl.py", line 31, in register
    if torch._C._dispatch_has_kernel_for_dispatch_key(self.qualname, "Meta"):
       ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
RuntimeError: operator _C::marlin_qqq_gemm does not exist

Is there a prebuilt docker image I could use? If not, could you please share how you built vllm to include marlin_qqq_gemm?

Thank you so much for your help!