When attempting to launch using a setup with the requirements_nocuda.txt, an error indicates a missing exllama dependency, which was previously not required for llama-cpp. Upon cloning the exllama repo into the repositories directory, a similar message is displayed, but with a different stacktrace.
Is there an existing issue for this?
[X] I have searched the existing issues
Reproduction
Follow the installation instructions for running off AMD and CPU
Follow steps 2 and 3 (building llama-cpp-python and ctransformers)
Attempt to launch the webui
Screenshot
No response
Logs
##Log without exllama
2023-09-17 20:34:09 WARNING:exllama module failed to import. Will attempt to import from repositories/.
2023-09-17 20:34:09 ERROR:Could not find repositories/exllama. Please ensure that exllama (https://github.com/turboderp/exllama) is cloned inside repositories/ and is up to date.
Traceback (most recent call last):
File "/path/text-generation-webui/modules/exllama.py", line 13, in <module>
from exllama.generator import ExLlamaGenerator
ModuleNotFoundError: No module named 'exllama'
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "/path/text-generation-webui/server.py", line 29, in <module>
from modules import (
File "/path/text-generation-webui/modules/ui_default.py", line 3, in <module>
from modules import logits, shared, ui, utils
File "/path/text-generation-webui/modules/logits.py", line 4, in <module>
from modules.exllama import ExllamaModel
File "/path/text-generation-webui/modules/exllama.py", line 22, in <module>
from generator import ExLlamaGenerator
ModuleNotFoundError: No module named 'generator'
##Log with exllama in repositories directory
2023-09-17 20:34:43 WARNING:exllama module failed to import. Will attempt to import from repositories/.
2023-09-17 20:34:43 ERROR:Could not find repositories/exllama. Please ensure that exllama (https://github.com/turboderp/exllama) is cloned inside repositories/ and is up to date.
Traceback (most recent call last):
File "/path/text-generation-webui/modules/exllama.py", line 13, in <module>
from exllama.generator import ExLlamaGenerator
ModuleNotFoundError: No module named 'exllama'
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "/path/text-generation-webui/server.py", line 29, in <module>
from modules import (
File "/path/text-generation-webui/modules/ui_default.py", line 3, in <module>
from modules import logits, shared, ui, utils
File "/path/text-generation-webui/modules/logits.py", line 4, in <module>
from modules.exllama import ExllamaModel
File "/path/text-generation-webui/modules/exllama.py", line 22, in <module>
from generator import ExLlamaGenerator
File "/path/text-generation-webui/repositories/exllama/generator.py", line 1, in <module>
import cuda_ext
File "/path/text-generation-webui/repositories/exllama/cuda_ext.py", line 43, in <module>
exllama_ext = load(
File "/home/user/.conda/envs/textgen2/lib/python3.10/site-packages/torch/utils/cpp_extension.py", line 1284, in load
return _jit_compile(
File "/home/user/.conda/envs/textgen2/lib/python3.10/site-packages/torch/utils/cpp_extension.py", line 1509, in _jit_compile
_write_ninja_file_and_build_library(
File "/home/user/.conda/envs/textgen2/lib/python3.10/site-packages/torch/utils/cpp_extension.py", line 1601, in _write_ninja_file_and_build_library
extra_ldflags = _prepare_ldflags(
File "/home/user/.conda/envs/textgen2/lib/python3.10/site-packages/torch/utils/cpp_extension.py", line 1699, in _prepare_ldflags
extra_ldflags.append(f'-L{_join_cuda_home("lib64")}')
File "/home/user/.conda/envs/textgen2/lib/python3.10/site-packages/torch/utils/cpp_extension.py", line 2223, in _join_cuda_home
raise EnvironmentError('CUDA_HOME environment variable is not set. '
OSError: CUDA_HOME environment variable is not set. Please set it to your CUDA install root.
Describe the bug
When attempting to launch using a setup with the requirements_nocuda.txt, an error indicates a missing exllama dependency, which was previously not required for llama-cpp. Upon cloning the exllama repo into the repositories directory, a similar message is displayed, but with a different stacktrace.
Is there an existing issue for this?
Reproduction
Screenshot
No response
Logs
System Info