Getting Started

Arewa's inference API is fully compatible with the OpenAI Chat Completions specification. If you're already using OpenAI, switching takes two lines: update your base_url and api_key.

1. Get your API key

Sign up at dashboard.arewa.ai and generate an API key from Settings → API Keys. Your key starts with sk-prod.

2. Make your first request

3. Choose a model

Replace "qwen3.5-9b-thinking" with any model ID from our model catalogue. All models use the same request schema; switching is a one-line change.

Key concepts

Workspaces isolate teams, billing, and access controls. Each workspace gets its own API key namespace and audit log.

Privacy tiers determine where your data is processed. Choose Cloud, Dedicated, BYO, or Sovereign at workspace creation; upgrade anytime without code changes.

Authentication

All API requests are authenticated with a Bearer token in the Authorization header. Keys are scoped to a workspace and start with sk-prod.

Get your key

1. Sign in at dashboard.arewa.ai
2. Navigate to Settings → API Keys
3. Click New API Key and name it for your environment

Send requests

Include your key in every API call:

Authorization: Bearer sk-prod-your-key-here

Store your key as an environment variable (AREWA_API_KEY) and never embed it in client-side code. Create one key per environment — development, staging, production — to track usage and rotate independently.

Key rotation

API keys can be revoked and reissued at any time from the dashboard without downtime. Revocation takes effect immediately across all endpoints.

Chat Completions

The primary endpoint for text and multimodal generation. Fully compatible with the OpenAI Chat Completions spec.

POST /v1/chat/completions

Required parameters

• model — model ID from the catalogue (e.g. qwen3.5-9b-thinking)
• messages — array of {"role", "content"} objects

Common optional parameters

• temperature — float 0–2, controls randomness. Default: 1.0
• max_tokens — integer, maximum tokens to generate
• stream — boolean, stream tokens via SSE. Default: false
• top_p — float 0–1, nucleus sampling threshold

Response

The response object contains an id for debugging, the model used, a choices array with the generated message, and a usage object with prompt, completion, and total token counts.

Streaming

Set stream: true to receive tokens as they are generated. Responses follow the server-sent events (SSE) format; each event is a JSON delta prefixed with data:. The stream terminates with data: [DONE].

Account Management API

The Account Management API lets you manage API keys, query usage metrics, and check your credit balance programmatically. All endpoints live under a separate base URL from the inference API.

Base URL:  https://api.arewa.ai/manage/v1

Authentication

Use your Arewa API key (sk-prod-...) as a Bearer token, the same key you use for inference calls. All endpoints also accept a JWT session token for dashboard-initiated requests.

Authorization: Bearer sk-prod-your-key-here

API Keys

Create and rotate keys programmatically. The full key value is returned only once at creation; store it securely immediately.

List keys

GET /api-keys
GET /api-keys?include_revoked=true   # include soft-deleted keys

Create a key

POST /api-keys
Content-Type: application/json

{ "name": "Production" }

→ 201 Created
{
  "id":         "key_abc123",
  "name":       "Production",
  "value":      "sk-prod-...",   ← shown once, store immediately
  "created_at": 1757892600
}

Revoke a key

DELETE /api-keys/{key_id}

→ 200 OK
{ "deleted": true }

Usage Metrics

Query token consumption, request counts, and cost over any time window. All timestamps are Unix seconds.

Single metric

GET /usage?metric=tokens&from=1756684800&to=1757548800&granularity=daily

# metric      → tokens | requests | cost
# granularity → daily | hourly | minute
# model       → model alias or "all"  (optional, default: all)
# api_key_id  → key ID or "all"       (optional, default: all)

All metrics in one call

GET /usage/summary?from=1756684800&to=1757548800&granularity=daily

→ 200 OK
{
  "tokens":   { "total": 4820000, "data": [...] },
  "requests": { "total": 1204,    "data": [...] },
  "cost":     { "total": 0.96,    "data": [...] }
}

Available Models

Returns all active models with per-million-token pricing. Use the id field as the model parameter in inference calls.

GET /models

→ 200 OK
[{
  "id":                 "qwen3.5-9b",
  "name":               "Qwen3.5-9B",
  "cost_mtoken_input":  0.15,
  "cost_mtoken_output": 0.60
}]

Credit Balance

Check your available spend in USD. Credits are consumed as inference requests complete and are deducted in real time.

GET /credits/balance

→ 200 OK
{ "balance": 12.50 }

Dashboard

The dashboard at dashboard.arewa.ai is your visual control panel; sign in with the same credentials you use for the API. It covers six areas:

Page	Path	What you can do
Models	/dashboard	Browse the full live model catalogue with real-time pricing. Click any model to see specs and copy its alias for use in API calls.
API Keys	/dashboard/api-keys	Create, name, and revoke inference keys (sk-prod-...). The full key value is shown only at creation; copy it immediately.
Usage	/dashboard/usage	Charts for token consumption, request count, and estimated cost. Filter by model or date range. Your current credit balance is shown at the top.
Billing	/dashboard/billing	Add or update a payment method and top up your credit balance. Minimum top-up is $5.
Playground	/dashboard/playground	Test any model interactively in the browser before wiring up an API integration. Requires an active API key and credit balance.
Profile	/dashboard/profile	View and update your account information and preferences.

Getting your first API key

1. Sign in at dashboard.arewa.ai
2. Complete onboarding if prompted, enter your role, company, and country
3. Open API Keys in the navigation
4. Click Create Key and give it a name (e.g. production, dev)
5. Copy the full key; it starts with sk-prod- and is shown only once
6. Store it as an environment variable and use it for inference calls

export AREWA_API_KEY="sk-prod-..."

curl https://api.arewa.ai/inference/v1/chat/completions \
  -H "Authorization: Bearer $AREWA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "model": "qwen3.5-9b", "messages": [{ "role": "user", "content": "Hello" }] }'