$ sudo dnf copr enable maxamillion/nvidia-openshell
$ sudo dnf install openshell-gateway openshell
$ systemctl --user start openshell-gateway.service
$ openshell gateway add https://localhost:17670 --local
$ export vertex_project="MY_ORGS_GCLOUD_PROJECT"
$ openshell settings set --global --key providers_v2_enabled --value true --yes
$ openshell provider create --name vertex-local --type google-vertex-ai --from-gcloud-adc --config VERTEX_AI_PROJECT_ID="${vertex_project}" --config VERTEX_AI_REGION=global
$ openshell inference set --provider vertex-local --model claude-sonnet-4-6
$ openshell sandbox create --name my-sandbox --provider vertex-local
$ ANTHROPIC_BASE_URL="https://inference.local" ANTHROPIC_API_KEY=unused claude --bare
$ ANTHROPIC_BASE_URL="https://inference.local/v1" ANTHROPIC_API_KEY=unused opencode
Red Hat: OpenShell + Podman + Fedora/RHEL Vertex AI Anthropic Quickstart
Red Hat has released a new integration enabling developers to run Anthropic's Claude Code and OpenCode AI coding assistants locally using OpenShell, Podman, and Fedora or RHEL systems with Vertex AI backend support. The setup allows users to create a local sandbox environment that connects to Google's Vertex AI for inference, bypassing the need for direct Anthropic API keys. Developers can install the necessary packages via COPR, configure the OpenShell gateway, and launch the AI assistants with environment variables pointing to the local inference endpoint.
Run your AI side-project on zahid.host
EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.