forked from phoenix-oss/llama-stack-mirror
# What does this PR do? - resolves issue: #1159 - Root cause: https://github.com/meta-llama/llama-stack/pull/980 forces `build_venv.sh` to install in a venv environment, which do not work on Colab notebook environment <img width="1004" alt="image" src="https://github.com/user-attachments/assets/1f9be409-5313-4926-b078-74e141cf29eb" /> ## This PR Use `UV_SYSTEM_PYTHON` to make sure dependencies are installed in current system environment. Which will be used in the Colab environment. ``` UV_SYSTEM_PYTHON=1 llama stack build --template together --image-type venv ``` ## Test Plan - Works in Colab environment <img width="621" alt="image" src="https://github.com/user-attachments/assets/ae93bc3d-e05a-44b9-bb21-fb88f29969b8" /> |
||
---|---|---|
.. | ||
routers | ||
server | ||
store | ||
ui | ||
utils | ||
__init__.py | ||
build.py | ||
build_conda_env.sh | ||
build_container.sh | ||
build_venv.sh | ||
client.py | ||
common.sh | ||
configure.py | ||
configure_container.sh | ||
datatypes.py | ||
distribution.py | ||
inspect.py | ||
library_client.py | ||
request_headers.py | ||
resolver.py | ||
stack.py | ||
start_conda_env.sh | ||
start_container.sh | ||
start_venv.sh |