Quick Start
Choose an option
| Option | Best for | What you need |
|---|---|---|
| XSI Control multi-tenant cloud xsicontrol.com | Using your own upstream keys without deploying anything | An email address and upstream keys |
| XSI hosted service xsirouter.com | Calling models directly without maintaining any upstream keys | An email address |
| Self-host XSI | Full control over data and operations | A Linux host, a domain, and upstream keys |
All three options expose the same API. In the examples below, replace only XSI_BASE_URL and XSI_API_KEY.
Option 1: Open an XSI Control tenant
XSI Control is the multi-tenant cloud edition of XSI. Each tenant has its own upstream key pool, routing configuration, pricing and account tree.
- Sign up with your email at a.xsicontrol.com and open the link in the activation email. The page shows your tenant owner API key.
- Sign in to the Admin Console (a.xsicontrol.com) with that key and add your own upstream keys (BYOK) under Model Providers, for example OpenAI in Level 1 and Anthropic in Level 2. XSI generates the matching
LEVEL_MAPPERentries. - In the Manage Console (m.xsicontrol.com), create sub-accounts for your team or customers and give them credit. Applications call with a sub-account's key.
- Set the environment variables:
export XSI_BASE_URL="https://api.xsicontrol.com"
export XSI_API_KEY="sk-Xvs..."Upstream keys are encrypted with your tenant key, which is wrapped by your API key; see Architecture.
Option 2: Self-host
git clone https://github.com/xabcstack/xsi.git && cd xsi
umask 077 && cp docker-compose.example.yml docker-compose.yml
sed -i "s/CHANGE_ME_POSTGRES_PASSWORD/$(openssl rand -hex 24)/g" docker-compose.yml
sed -i "s/CHANGE_ME_ROOT_KEY/sk-Xvs$(openssl rand -hex 24 | cut -c1-45)/" docker-compose.yml
# Edit docker-compose.yml: domains, root email and SMTP
docker compose up -dThen:
- Set up an HTTPS reverse proxy for the API and both consoles. See Reverse Proxy.
- Sign in to the Admin Console with the root key (the
sk-Xvs...value inINIT_USERS) and add upstream keys under Model Providers. - In the Manage Console, create a sub-account, allocate credit to it, and call models with the sub-account's key.
For the complete procedure, see Docker Compose Deployment.
export XSI_BASE_URL="https://api.example.com"
export XSI_API_KEY="<sub-account key>"Send your first request
The model names in these examples must be configured in the price list, and your account must be allowed to use them. Check with GET /v1/models or GET /x-pricing.
OpenAI Responses
curl "$XSI_BASE_URL/v1/responses" \
-H "Authorization: Bearer $XSI_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model": "gpt-6.2-sol", "input": "Introduce yourself in one sentence"}'OpenAI Chat Completions
curl "$XSI_BASE_URL/v1/chat/completions" \
-H "Authorization: Bearer $XSI_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model": "gpt-6.2-sol", "messages": [{"role": "user", "content": "Introduce yourself in one sentence"}]}'Anthropic Messages
curl "$XSI_BASE_URL/v1/messages" \
-H "x-api-key: $XSI_API_KEY" \
-H "anthropic-version: 2023-06-01" \
-H "Content-Type: application/json" \
-d '{"model": "claude-sonnet-5-5", "max_tokens": 1024, "messages": [{"role": "user", "content": "Introduce yourself in one sentence"}]}'Gemini
curl "$XSI_BASE_URL/v1beta/models/$GEMINI_MODEL:generateContent" \
-H "x-goog-api-key: $XSI_API_KEY" \
-H "Content-Type: application/json" \
-d '{"contents": [{"parts": [{"text": "Introduce yourself in one sentence"}]}]}'DeepSeek
curl "$XSI_BASE_URL/v1/chat/completions" \
-H "Authorization: Bearer $XSI_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model": "deepseek-flash", "messages": [{"role": "user", "content": "Introduce yourself in one sentence"}]}'Check usage
curl "$XSI_BASE_URL/dashboard/live" -H "Authorization: Bearer $XSI_API_KEY"You can also view live usage, billing, and per-model details in the Manage Console.
Next steps
- SDK Integration: Python, Node.js, and other clients
- Router API: all model endpoints
- Account Hierarchy: create sub-accounts for your team or customers
- Routing and Failover: configure multiple upstreams and disaster recovery