Skip to main content
Seattle Compute Co.

Docs · Sheet 01

Quickstart

Go from zero to a dedicated inference endpoint in five minutes.

1. Install & auth

Create an account in the dashboard. You’ll receive an API key shown once — store it securely. All API calls use the X-API-Key header.

shellbash
export SC_API_KEY="sc_live_..."
export SC_BASE="https://api.seattlecompute.com/v1"

2. Deploy a model

Create a deployment by specifying a model id, priority, and projected hours per day. The optimizer picks the GPU and engine for you.

deploy.shbash
curl -X POST "$SC_BASE/deployments" \
  -H "X-API-Key: $SC_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model_source": "huggingface",
    "model_id": "meta-llama/Llama-3.1-8B-Instruct",
    "priority": "balanced",
    "hours_per_day": 24,
    "precision": "fp16"
  }'

3. Call your endpoint

Every deployment exposes an OpenAI-compatible chat completion endpoint.

inference.shbash
curl "$ENDPOINT_URL/v1/chat/completions" \
  -H "Authorization: Bearer $SC_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "llama-3.1-8b-instruct",
    "messages": [{"role": "user", "content": "Hello from Seattle."}]
  }'

4. List deployments

list.shbash
curl "$SC_BASE/deployments" -H "X-API-Key: $SC_API_KEY"

5. Custom domains

Register your domain, point the CNAME we give you, and attach to any deployment. TLS is provisioned automatically.

domain.shbash
curl -X POST "$SC_BASE/domains" \
  -H "X-API-Key: $SC_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"domain": "llm.yourcompany.com"}'

6. Next steps