Description
Hello.
When executing “chat with rtx” the following lines of code and error appear and it doesn’t start, previously install chatrtx and everything is fine!!
Environment
I have an RTX-4090 and core i9 14900 HX
Error
[TensorRT-LLM] TensorRT-LLM version: 0.9.0
Traceback (most recent call last):
File “C:\Users\HM\AppData\Local\NVIDIA\ChatWithRTX\RAG\trt-llm-rag-windows-main\app.py”, line 97, in
llm = TrtLlmAPI(
File “C:\Users\HM\AppData\Local\NVIDIA\ChatWithRTX\RAG\trt-llm-rag-windows-main\trt_llama_api.py”, line 123, in init
self._model = runner_cls.from_dir(**runner_kwargs)
File “C:\Users\HM\AppData\Local\NVIDIA\ChatRTX\env_nvd_rag\lib\site-packages\tensorrt_llm\runtime\model_runner.py”, line 483, in from_dir
model_config, other_config = read_config(config_path)
File “C:\Users\HM\AppData\Local\NVIDIA\ChatRTX\env_nvd_rag\lib\site-packages\tensorrt_llm\runtime\model_runner.py”, line 78, in read_config
return _builder_to_model_config(config)
File “C:\Users\HM\AppData\Local\NVIDIA\ChatRTX\env_nvd_rag\lib\site-packages\tensorrt_llm\runtime\model_runner.py”, line 132, in _builder_to_model_config
mamba_conv1d_plugin = bool(plugin_config[‘mamba_conv1d_plugin’])
KeyError: 'mamba_conv1d_plugin