Remove build and installation of whisper.cpp, and installation of ffmpeg.
Rename build_llama_and_whisper.sh to build_llama.sh.
Update Containerfiles to reference new script name.
Consolidate management of cmake args in build_llama.sh.
Remove references to whisper-server in various locations.
Signed-off-by: Mike Bonnet <mikeb@redhat.com>
The entrypoint images are used by AI Lab, which pins to older digests and
has no need to update. Stop building and releasing new, unused versions.
Signed-off-by: Mike Bonnet <mikeb@redhat.com>
Since RAG has been moved into its own images, no python runtime environment is
required in the inference engine images.
Signed-off-by: Mike Bonnet <mikeb@redhat.com>
assisted by cursor/claude.
Signed-off-by: Jeff Ligon <jligon@redhat.com>
going with gemini's python_check(with the double equal) for this one.
Signed-off-by: Jeff Ligon <jligon@redhat.com>
changing to `MYPIP ?= pipx`
Signed-off-by: Jeff Ligon <jligon@redhat.com>
fixing uv pipx
Signed-off-by: Jeff Ligon <jligon@redhat.com>
taking it back to pip to fix ci
Signed-off-by: Jeff Ligon <jligon@redhat.com>
A previous commit changed the second argument to add_rag() from the image name to the
full repo path. Update the case statement accordingly, so the "GPU" variable is set correctly.
The "cuda" directory is no longer available on download.pytorch.org. When building for cuda,
pull wheels from the "cu128" directory, which contains binaries built for CUDA 12.8.
When building rocm* images, download binaries from the "rocm6.3" directory, which are built
for ROCm 6.3.
Signed-off-by: Mike Bonnet <mikeb@redhat.com>
Move the Containerfiles for the entrypoint and rag images out of container_build.sh and into their
own files. This is necessary so they can be built with Konflux.
Signed-off-by: Mike Bonnet <mikeb@redhat.com>
Add a new "bats" container which is configured to run the bats tests.
The container supports running the standard bats test suite
(container-in-container) as well as the "--nocontainer" tests.
Add two new Makefile targets for running the bats container via podman.
Signed-off-by: Mike Bonnet <mikeb@redhat.com>
This commit adds TMT test jobs triggered via Packit that fetches an
instance with NVIDIA GPU, specified in `plans/no-rpm.fmf`, and can be
verified in the gpu_info test result.
In addition, system tests (nocontainer), validate, and unit tests are
also triggered via TMT.
Fixes: #1054
TODO:
1. Enable bats-docker tests
2. Resolve f41 validate test failures
Signed-off-by: Lokesh Mandvekar <lsm5@fedoraproject.org>
Copy the current checkout of the ramalama repo into the containers and use that for installation.
This removes the need for an extra checkout of the ramalama repo, and is consistent with the build
process used by container_build.sh (which used a bind-mount rather than a copy).
This keeps the version of ramalama in sync with the Containerfiles, and makes testing and CI more
useful.
Signed-off-by: Mike Bonnet <mikeb@redhat.com>
This image will just run ramalama inside of a container and
requires the user to leak the podman-socket into the container.
It will use Podman-remote for all of its actions.
Requested by the Podman Desktop team.
Fixes: https://github.com/containers/ramalama/issues/837
Signed-off-by: Daniel J Walsh <dwalsh@redhat.com>
Remove pragmatic, and move to using local implementation
until llama-stack version is ready.
python3 container-images/scripts/doc2rag.py --help
usage: docling [-h] target source [source ...]
process source files into RAG vector database
positional arguments:
target
source
options:
-h, --help show this help message and exit
ramalama rag should be using accelerated images.
Signed-off-by: Daniel J Walsh <dwalsh@redhat.com>
This PR just installs the python requirements needed to play with the
rag_framework.py file.
I have not added the docling support yet, since that would swell the
size of the images. Will add that in a separate PR.
Also remove pragmatic and begin conversion to new rag tooling.
Signed-off-by: Brian <bmahabir@bu.edu>
Signed-off-by: Daniel J Walsh <dwalsh@redhat.com>
Install podman-remote and ramalama so we can use ramalama
from within a container.
$ podman run --env XDG_RUNTIME_DIR -v $HOME:$HOME -v /run/user:/run/user --userns=keep-id -ti --privileged quay.io/ramalama/ramalama ramalama run granite
Signed-off-by: Daniel J Walsh <dwalsh@redhat.com>
All builds inherit just from the OS image. Helps save space as the
derived images just overwrite llama.cpp and whisper.cpp which is
wasteful.
Signed-off-by: Eric Curtin <ecurtin@redhat.com>
Update CUDA base image from version to 12.6.3 for both builder
and runtime stages. Update UBI base image to 9.5. Update ROCm
version in to 6.3.1.
Signed-off-by: Eric Curtin <ecurtin@redhat.com>
- Modified `Makefile` to pass the 'build' command to `container_build.sh`.
- Refactored `container_build.sh` to:
- Use `case` statements for command handling.
- Add `determine_platform` function to determine the platform.
- Add `parse_arguments` function for better argument parsing.
- Add `process_all_targets` function to handle all container images.
- Add `print_usage` function to display usage information.
- Enhance the `main` function to use the new functions and handle errors.
Signed-off-by: Eric Curtin <ecurtin@redhat.com>
Add this image, we also want to see the size of this image in the
build.
Co-authored-by: Steffen Roecker <sroecker@redhat.com>
Signed-off-by: Eric Curtin <ecurtin@redhat.com>
