Analytics cookies only (Google Analytics, PostHog). No ad networks, no cross-site tracking. See our privacy policy.
A workspace for building, deploying and monitoring AI agents and workflows. You assemble agents on a visual flow editor, connect them to a long list of integrations (Slack, Stripe, Notion, Salesforce, Supabase and many more) and to MCP servers, and run them from the app. It works with hosted models and with local ones through Ollama or vLLM. The self-hosted stack is Next.js on Bun with PostgreSQL and pgvector, Redis for pub/sub and caches, a Socket.io realtime service for the editor and a one-shot migrations job. Features such as single sign-on, access control groups and SCIM sit in an ee/ directory under a separate Enterprise license; the published image contains that code, which stays off unless the enterprise feature flags are set.
Runs on a VPS from $24/mo (Kamatera 8GB, 8 GB RAM) · prices checked Nov 2026
Fits home hardware (877 MB measured idle, RAM only): Raspberry Pi 4 (4 GB) (arm64 untested), Raspberry Pi 5 (8 GB) (arm64 untested), Repurposed laptop (8 GB), Intel N100 mini PC (16 GB), AMD Ryzen mini PC (32 GB).
Pick Sim when you want to build agents and workflows on a visual flow editor and connect them to a large set of ready-made integrations and MCP servers, on infrastructure you control. It is a Next.js and Bun app with PostgreSQL and pgvector, Redis, a Socket.io realtime service and a migrations job, supports hosted models and local ones through Ollama or vLLM, and the core is Apache-2.0 (the top-level licence file). Be clear about one thing: the repository also contains an `apps/sim/ee/` directory, with its own Sim Enterprise License, for features such as SSO, access-control groups and SCIM; that licence allows development and evaluation and requires an Enterprise subscription for production use of those features. The published image contains the `ee/` code, which is off unless the enterprise feature flags are set. Our snippet adapts the project's `docker-compose.prod.yml`, pins `simstudio`, `realtime` and `migrations` to `v0.9.48` (released 2026-11-08, amd64 and arm64), pins `pgvector/pgvector:0.8.7-pg17` and `redis:7.4.12-alpine` in place of floating tags, generates the secrets into `.env`, and leaves out the optional cron service. The self-hosting docs list 12 GiB as the lowest "Evaluation example" while saying these are "not enforced minimums", and the compose file gives the app a limit of 8 GB; our 4096 MB floor is well below that example, and we measured 877 MB idle with no workflows. Among the catalog's agent platforms, Dify, Flowise and Langflow are also visual builders for LLM applications and agents; Sim lists 270+ integrations and runs as five containers in our snippet, and it ships about one release a day, sometimes more (v0.9.45 to v0.9.48 between 2026-11-06 and 2026-11-08), so expect fast change.
Paid link — we earn a commission if you shop through it.
Run these commands on your server:
# Sim — AI agent workflow builder — adapted from the project's docker-compose.prod.yml, pinned to v0.9.48 (app, realtime and migrations share one schema and must use the same tag); the optional cron service is left out
echo "BETTER_AUTH_SECRET=$(openssl rand -hex 32)" > .env
echo "ENCRYPTION_KEY=$(openssl rand -hex 32)" >> .env
echo "INTERNAL_API_SECRET=$(openssl rand -hex 32)" >> .env
echo "API_ENCRYPTION_KEY=$(openssl rand -hex 32)" >> .env
echo "POSTGRES_PASSWORD=$(openssl rand -hex 24)" >> .env
cat > docker-compose.yml <<'EOF'
services:
simstudio:
image: ghcr.io/simstudioai/simstudio:v0.9.48
restart: unless-stopped
ports:
- 3000:3000
env_file: .env
environment:
NODE_ENV: production
DATABASE_URL: postgresql://postgres:${POSTGRES_PASSWORD}@db:5432/simstudio
BETTER_AUTH_URL: http://localhost:3000
NEXT_PUBLIC_APP_URL: http://localhost:3000
REDIS_URL: redis://redis:6379
SOCKET_SERVER_URL: http://realtime:3002
depends_on:
db:
condition: service_healthy
redis:
condition: service_healthy
migrations:
condition: service_completed_successfully
realtime:
condition: service_healthy
healthcheck:
test: ["CMD", "curl", "-fsS", "http://127.0.0.1:3000/api/health"]
interval: 30s
timeout: 5s
retries: 5
start_period: 30s
realtime:
image: ghcr.io/simstudioai/realtime:v0.9.48
restart: unless-stopped
ports:
- 3002:3002
env_file: .env
environment:
NODE_ENV: production
DATABASE_URL: postgresql://postgres:${POSTGRES_PASSWORD}@db:5432/simstudio
NEXT_PUBLIC_APP_URL: http://localhost:3000
BETTER_AUTH_URL: http://localhost:3000
REDIS_URL: redis://redis:6379
depends_on:
db:
condition: service_healthy
redis:
condition: service_healthy
healthcheck:
test: ["CMD", "curl", "-fsS", "http://127.0.0.1:3002/health"]
interval: 30s
timeout: 5s
retries: 5
start_period: 10s
migrations:
image: ghcr.io/simstudioai/migrations:v0.9.48
working_dir: /app/packages/db
environment:
DATABASE_URL: postgresql://postgres:${POSTGRES_PASSWORD}@db:5432/simstudio
ENCRYPTION_KEY: ${ENCRYPTION_KEY}
depends_on:
db:
condition: service_healthy
command: ["bun", "run", "db:migrate"]
restart: "no"
redis:
image: redis:7.4.12-alpine
restart: unless-stopped
healthcheck:
test: ["CMD", "redis-cli", "ping"]
interval: 5s
timeout: 5s
retries: 5
db:
image: pgvector/pgvector:0.8.7-pg17
restart: unless-stopped
environment:
POSTGRES_USER: postgres
POSTGRES_PASSWORD: ${POSTGRES_PASSWORD}
POSTGRES_DB: simstudio
volumes:
- postgres_data:/var/lib/postgresql/data
healthcheck:
test: ["CMD-SHELL", "pg_isready -U postgres"]
interval: 5s
timeout: 5s
retries: 5
volumes:
postgres_data:
EOF
docker compose up -d
timeout 600 bash -c 'until curl -fsS -o /dev/null http://localhost:3000/api/health; do sleep 3; done'
# → http://SERVER_IP:3000 — the Sim sign-up and login page. Before using it beyond localhost, edit NEXT_PUBLIC_APP_URL and BETTER_AUTH_URL in both services (and set NEXT_PUBLIC_SOCKET_URL for the editor's live connection)Sim's README, Docker docs and compose file flag these before you rely on it:
Sim used 877 MB of RAM at idle on GCP e2-standard-2 · Ubuntu 26.04 LTS · Docker 29.9.3 (first boot, measured Nov 2026). We list a minimum of 4096 MB.
The Sim install snippet is marked verified on Ubuntu 26.04 (Nov 2026).