Failed: Loading model deepseek-r1-distill-qwen-14b-trtrtx-gpu:1

[ddobric]

Hi all,

trying to load some models with the following error:

PS C:\Users\me> foundry model run deepseek-r1-14b
Model deepseek-r1-distill-qwen-14b-trtrtx-gpu:1 was found in the local cache.
🕕 Loading model... [20:42:42 ERR] Failed loading model:deepseek-r1-distill-qwen-14b-trtrtx-gpu:1

Exception: Failed: Loading model deepseek-r1-distill-qwen-14b-trtrtx-gpu:1 from http://127.0.0.1:62676/openai/load/deepseek-r1-distill-qwen-14b-trtrtx-gpu:1?ttl=600
Internal Server Error
Failed loading model deepseek-r1-distill-qwen-14b-trtrtx-gpu:1
NvTensorRTRTX EP failed to deserialize engine for fused node: NvTensorRTRTXExecutionProvider_TRTKernel_graph_main_graph_15183841921133510762_0_0

PS C:\Users\me> foundry model run qwen2.5-14b
Model qwen2.5-14b-instruct-trtrtx-gpu:2 was found in the local cache.
🕗 Loading model... [21:45:47 ERR] Failed loading model:qwen2.5-14b-instruct-trtrtx-gpu:2
Exception: Failed: Loading model qwen2.5-14b-instruct-trtrtx-gpu:2 from http://127.0.0.1:62676/openai/load/qwen2.5-14b-instruct-trtrtx-gpu:2?ttl=600
Internal Server Error
Failed loading model qwen2.5-14b-instruct-trtrtx-gpu:2
NvTensorRTRTX EP failed to deserialize engine for fused node: NvTensorRTRTXExecutionProvider_TRTKernel_graph_main_graph_12702861025774906949_0_0

Is there any hint, how to analyse such issues and fix the problem?

Thanks

[metang]

what's the hardware spec. This could be an issue that not enough VRAM memory to load the model.