vLLM Studio

Unified local AI workstation for model lifecycle, chat/agent workflows, orchestration, observability, and remote deployment.

Release: v1.13.0

This release consolidates major repo changes currently in the tree, including:

• OpenAI proxy activation policy controls for load_if_idle and switch_on_request
• lifecycle-aware run aborts when model eviction happens
• SSE run stream termination fixes across backend and frontend
• Daytona toolbox reliability hardening (route fallback, quota cleanup retry, alias-safe command parsing)
• dashboard launch-state cleanup improvements
• expanded controller/daytona test coverage for run termination and toolbox behavior

Docs

• Overview: docs/README.md
• Setup and deployment: setup/README.md
• Environment variables: docs/environment.md

Repository layout

• controller/: Bun/Hono backend, orchestration, chat runtime, lifecycle, metrics
• frontend/: Next.js app, chat UI, proxy endpoints, client state
• cli/: Bun CLI for controller access
• swift-client/: iOS/macOS client (XcodeGen)
• shared/: shared types/contracts
• config/: runtime and integration configs
• docs/: documentation index and environment notes
• scripts/: operational scripts (deployment + controller daemon helpers)
• docker-compose.yml: full stack service definitions
• scripts/daemon-*.sh: start/status/stop helpers for background controller runs

Quick start

1. Controller (local):

cd controller
npx tsc --noEmit
bun test
bun src/main.ts

2. Frontend:

cd frontend
npm run test
npm run lint
npm run build
npm run dev

3. Full stack with Docker (controller + frontend + infra):

docker compose up -d --build controller frontend

4. Run controller as a background daemon:

./scripts/daemon-start.sh
./scripts/daemon-status.sh
./scripts/daemon-stop.sh

Health checks

curl -sS http://localhost:8080/health
curl -I http://localhost:3000

API docs

• http://localhost:8080/api/docs
• http://localhost:8080/api/spec

Setup guide

See setup/README.md for complete setup, deployment, and verification instructions.

Branching and release workflow

• Development branch: dev
• Production integration branch: main
• Release tags: vX.Y.Z

For this release:

• merge release work into main and dev
• tag v1.13.0
• create a new post-release working branch