Skip to content

build(deps): bump llama-cpp-python from 0.2.63 to 0.2.65 #59

build(deps): bump llama-cpp-python from 0.2.63 to 0.2.65

build(deps): bump llama-cpp-python from 0.2.63 to 0.2.65 #59

Workflow file for this run

name: Build and Push
on:
push:
branches: ["main"]
paths-ignore:
- "**.md"
pull_request:
paths-ignore:
- "**.md"
workflow_dispatch:
env:
QUAY_REPO: "quay.io/dtrifiro/llama-cpp-python-serving"
CI: true
DOCKER_BUILDKIT: 1
FORCE_COLOR: 1
jobs:
build-and-push:
name: Build and Push Image
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v4
- name: "Setup Docker Buildx"
uses: docker/setup-buildx-action@v3
- name: "Build"
run: |
date="$(date +%Y%m%d%H%M)"
rev="$(git rev-parse --short HEAD)"
docker build -t "${QUAY_REPO}:latest" -t "${QUAY_REPO}:${rev}" -t "${QUAY_REPO}:${date}" .
- name: "Build CUDA image"
run: |
date="$(date +%Y%m%d%H%M)"
rev="$(git rev-parse --short HEAD)"
docker build -t "${QUAY_REPO}:latest-cuda" -t "${QUAY_REPO}:${rev}-cuda" -t "${QUAY_REPO}:${date}-cuda" -f Dockerfile.cuda .
- name: "Push to quay"
if: github.ref == 'refs/heads/main' && github.event_name != 'pull_request'
run: |
docker login -u="${{secrets.QUAY_USERNAME}}" -p="${{secrets.QUAY_TOKEN}}" quay.io
docker push --all-tags "${QUAY_REPO}"
smoke-test:
name: "Smoke test"
needs: build-and-push
if: github.ref == 'refs/heads/main' && github.event_name != 'pull_request'
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v4
- uses: dtrifiro/[email protected]
- name: Setup Model
run: |
bash scripts/setup_model.sh
- name: Deploy InferenceService/ServingRuntime
run: |
bash scripts/deploy.sh cpu
- name: Perform sample inference
run: |
ISVC_HOST=$(kubectl get isvc llama-cpp-python -o jsonpath='{.status.components.predictor.url}' | cut -d / -f 3-)
# Add the ISVC_HOST to /etc/hosts since we can't resolve it
echo "127.0.0.1 ${ISVC_HOST}" | sudo tee -a /etc/hosts
bash scripts/inference_simple.sh