Closed teis-e closed 3 weeks ago
I have met the bug, too
I never found the fix.
What did helped is to install it in a complete new virtual environment with specificity python 3.8. Then the pip install errors disappeared. But sadly i still ended up with some errors that i never got to fix.
Is there an existing issue for this bug?
🐛 Describe the bug
~/ColossalAI/examples/inference/llama$ colossalai run --nproc_per_node 1 llama_generation.py -m "~/llama8b/TensorRT-LLM/Meta-Llama-3-8B-Instruct" --max_length 80
Output:
Than after doing this:
cd ColossalAI
pip install .
This is the error during same run:
Environment
Python 3.10.14 Torch 2.3.1+cu121 CUDA Version: 12.4 nvcc: NVIDIA (R) Cuda compiler driver Copyright (c) 2005-2024 NVIDIA Corporation Built on Thu_Mar_28_02:18:24_PDT_2024 Cuda compilation tools, release 12.4, V12.4.131 Build cuda_12.4.r12.4/compiler.34097967_0