failed to build the serialized network for a simple model with only Or and Cast: Internal Error (Could not find any implementation for node)

Description

For the following simple onnx model,

the results produced by onnxruntime are as follows:

 [array([[[[1.]],

        [[1.]],

        [[1.]]]], dtype=float32)]

However, when I run it using tensorrt, an internal error occurred as follows:

Internal Error: MyelinCheckException: host_instr.cpp:2806: CHECK(0) failed. 
[07/06/2025-10:30:33] [TRT] [E] Error Code: 9: Skipping tactic 0x0000000000000000 due to exception [myelin_graph.h:attachExceptionMsgToGraph:1139] MyelinCheckException: host_instr.cpp:2806: CHECK(0) failed. 
[07/06/2025-10:30:33] [TRT] [E] IBuilder::buildSerializedNetwork: Error Code 10: Internal Error (Could not find any implementation for node {ForeignNode[or_constant_out + ONNXTRT_Broadcast...node_of_output]}.)

Environment

TensorRT Version: 10.12.0.36

NVIDIA GPU: GeForce RTX 3080

NVIDIA Driver Version: 535.183.01

CUDA Version: 12.2

CUDNN Version: none

Operating System: ubuntu 20.04

Python Version (if applicable): 3.12.9

Relevant Files

Model link:

Steps To Reproduce

This issue can be reproduced by the following code with the model in the attachment.

from typing import Dict, List, Literal, Optional
import sys
import os

import numpy as np
import onnx
import onnxruntime

import tensorrt as trt

import argparse
import pickle

def test():
    onnx_model = onnx.load('222.onnx')
    
    with open("inputs.pkl", "rb") as fp:
        inputs = pickle.load(fp)

    try:
        ort_session = onnxruntime.InferenceSession(
            onnx_model.SerializeToString(), providers=["CPUExecutionProvider"]
        )
        ort_output = ort_session.run([], inputs)
    except Exception as e:
        print(e)
        print("This model cannot be executed by onnxruntime!")
        sys.exit(1)
    
    print("ONNXRuntime:\n", ort_output)
    
    #--------------------------------------------------------
        
    trt_logger = trt.Logger(trt.Logger.WARNING)
    trt.init_libnvinfer_plugins(trt_logger, '')
    builder = trt.Builder(trt_logger)
    #network = builder.create_network()
    network = builder.create_network(flags=1 << int(trt.NetworkDefinitionCreationFlag.EXPLICIT_BATCH))

    parser = trt.OnnxParser(network, trt_logger)
    with open('222.onnx', 'rb') as model_file:
        if not parser.parse(model_file.read()):
            for error in range(parser.num_errors):
                print(parser.get_error(error))
            sys.exit(1)
    
    config = builder.create_builder_config()
    serialized_engine = builder.build_serialized_network(network, config)
    
if __name__ == "__main__":
    test()
    

testcase.zip

Commands or scripts:

Have you tried the latest release?: yes

Can this model run on other frameworks? For example run ONNX model with ONNXRuntime (polygraphy run <model.onnx> --onnxrt): the mode can be executed by onnxruntime.

Start by running the Python reproduction with testcase.zip, including 222.onnx and inputs.pkl, and compare the ONNXRuntime result with TensorRT's build_serialized_network failure. Inspect the parser and builder output around the Or and Cast nodes. Done means identifying why no implementation is selected or documenting a confirmed fix with a successful serialized network build.