API Reference

Send your first request in under a minute. The W1W API is fully compatible with the OpenAI chat completions format.

curl https://we1web-api.fly.dev/api/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "w1w-swarm",
    "messages": [{"role": "user", "content": "Hello"}],
    "stream": false
  }'

{
  "id": "chatcmpl-w1w-abc123",
  "object": "chat.completion",
  "model": "w1w-swarm",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Hello! How can I help you today?"
      },
      "finish_reason": "stop"
    }
  ],
  "_w1w": {
    "source": "swarm",
    "fallbackUsed": false,
    "durationMs": 342,
    "verificationStatus": "unverified",
    "nodesUsed": ["node-abc123"]
  }
}

Authenticate requests with an API key from your dashboard. Free tier requests work without authentication but are rate-limited.

Authorization: Bearer w1w_sk_your_api_key_here

GET /api/v1/chat/completions?api_key=w1w_sk_your_api_key_here

W1W is a drop-in replacement for the OpenAI API. Point any OpenAI SDK at our base URL and everything works — same request format, same response structure, plus the extra _w1w provenance field.

from openai import OpenAI

client = OpenAI(
    base_url="https://we1web-api.fly.dev/api/v1",
    api_key="YOUR_W1W_API_KEY"
)

# Works exactly like OpenAI
response = client.chat.completions.create(
    model="w1w-swarm",
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "Explain quantum computing in simple terms"}
    ],
    temperature=0.7,
    max_tokens=1024
)

print(response.choices[0].message.content)

# Access W1W provenance via the raw response
raw = response.model_extra  # or response._raw_response
print(f"Source: {raw.get('_w1w', {}).get('source')}")

Set stream: true to receive Server-Sent Events (SSE). Each chunk contains a delta of the response as it is generated.

curl https://we1web-api.fly.dev/api/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "w1w-swarm",
    "messages": [{"role": "user", "content": "Write a haiku"}],
    "stream": true
  }'

data: {"id":"chatcmpl-w1w-abc123","object":"chat.completion.chunk","choices":[{"index":0,"delta":{"content":"Hello"},"finish_reason":null}]}

data: {"id":"chatcmpl-w1w-abc123","object":"chat.completion.chunk","choices":[{"index":0,"delta":{"content":"!"},"finish_reason":null}]}

data: {"id":"chatcmpl-w1w-abc123","object":"chat.completion.chunk","choices":[{"index":0,"delta":{},"finish_reason":"stop"}],"_w1w":{"source":"swarm","fallbackUsed":false,"durationMs":215,"verificationStatus":"unverified","nodesUsed":["node-abc123"]}}

data: [DONE]

There are no monthly plans. Limits unlock permanently from your cumulative prepaid credit spend, and never lapse or downgrade. Every tier includes full provenance metadata and swarm-first routing, and rate limit headers are included in every response.

X-RateLimit-Limit: 30
X-RateLimit-Remaining: 27
X-RateLimit-Reset: 1714531200
Retry-After: 60

Every W1W response includes a _w1w object with provenance metadata: source, fallback status, latency, verification status, and the IDs of the nodes that served the request — plus a signed receipt where available. Verification is sampled server-side re-checking, not caller-side cryptographic proof, and we say so. That honesty is what makes W1W different from centralized APIs.

"_w1w": {
  "source": "swarm",
  "fallbackUsed": false,
  "durationMs": 342,
  "verificationStatus": "unverified",
  "nodesUsed": ["node-abc123"]
}

"_w1w": {
  "source": "gemini",
  "fallbackUsed": true,
  "durationMs": 1205,
  "verificationStatus": "centralized",
  "nodesUsed": []
}

Try the API right here. Free tier works without an API key. Paste your key for higher limits.