ConsoleAPI Console

Build with deMac

Connect with the OpenAI SDK, choose a model, and send your first request.

Base URL

https://api.demac.example/v1

Quick start

Install your preferred SDK and point it to deMac. Create an API key below, then add it to the example.

# No installation needed
export DEMAC_API_KEY="<YOUR_API_KEY>"
export DEMAC_BASE_URL="https://api.demac.example/v1"

Replace the model placeholder with an ID returned by /v1/models. Availability depends on connected providers.

API Keys

All keys draw from your shared account balance. A key's spend cap is a sub-limit on that balance, not extra funds.

This console uses one active key (saved in this browser) for its own chat and test calls. It's provisioned automatically; you can also point it at a new key below.

Connect your wallet to create and manage API keys.

Endpoint reference

Request formats, responses, and authentication requirements.

8 endpoints

Requires an Authorization: Bearer <api_key> header.

Request

{
  "model": "<model-id-from-/v1/models>",
  "messages": [
    {"role": "system", "content": "You are a helpful assistant."},
    {"role": "user", "content": "Hello!"}
  ],
  "stream": true,
  "max_tokens": 1024
}

Response

{
  "id": "chatcmpl-...",
  "object": "chat.completion.chunk",
  "model": "<model-id-from-/v1/models>",
  "choices": [{
    "index": 0,
    "delta": {"role": "assistant", "content": "Hello"},
    "finish_reason": null
  }]
}

Supports streaming (SSE) and non-streaming responses. All prompts are end-to-end encrypted. Response headers include provider attestation metadata (x-provider-attested, x-provider-trust-level, x-provider-chip).

Requires an Authorization: Bearer <api_key> header.

Request

{
  "model": "<model-id-from-/v1/models>",
  "input": "Explain how decentralized inference works.",
  "stream": true,
  "max_output_tokens": 1024
}

Response

{
  "id": "resp_...",
  "object": "response",
  "model": "<model-id-from-/v1/models>",
  "output": [{
    "type": "message",
    "role": "assistant",
    "content": [{
      "type": "output_text",
      "text": "Decentralized inference distributes..."
    }]
  }],
  "usage": {
    "input_tokens": 12,
    "output_tokens": 256
  }
}

OpenAI Responses API format. Accepts 'input' (string or array) instead of 'messages'. Uses input_tokens/output_tokens for usage. Supports streaming. Same routing, encryption, and billing as chat completions.

Requires an Authorization: Bearer <api_key> header.

Response

{
  "data": [
    {
      "id": "<model-id-from-/v1/models>",
      "object": "model",
      "name": "Gemma 4 26B",
      "hugging_face_id": "<model-id-from-/v1/models>",
      "created": 1735689600,
      "description": "Balanced general-purpose model.",
      "input_modalities": ["text"],
      "output_modalities": ["text"],
      "quantization": "int8",
      "context_length": 262144,
      "max_output_length": 16384,
      "pricing": {
        "prompt": "0.00000005",
        "completion": "0.0000002",
        "image": "0",
        "request": "0",
        "input_cache_read": "0"
      },
      "supported_sampling_parameters": ["temperature", "top_p", "top_k", "stop", "seed", "max_tokens"],
      "supported_features": ["tools", "reasoning"],
      "metadata": {
        "model_type": "chat",
        "provider_count": 2,
        "trust_level": "hardware",
        "display_name": "Gemma 4 26B"
      }
    }
  ]
}

OpenAI-compatible model list. Top-level fields include per-token USD pricing strings, modalities, and supported features. deMac-native fields such as trust_level and provider_count live under metadata.

Response

{
  "providers": [{
    "provider_id": "...",
    "chip_name": "Apple M4 Max",
    "hardware_model": "Mac16,1",
    "trust_level": "hardware",
    "status": "online",
    "secure_enclave": true,
    "sip_enabled": true,
    "secure_boot_enabled": true,
    "mdm_verified": true,
    "mda_verified": true,
    "se_public_key": "..."
  }]
}

Publicly accessible — no authentication required. Hardware serials, UDIDs, and raw Apple certificate chains stay private to the provider and coordinator.

Requires an Authorization: Bearer <api_key> header.

Response

{
  "usage": [
    {
      "request_id": "...",
      "model": "<model-id-from-/v1/models>",
      "prompt_tokens": 150,
      "completion_tokens": 500,
      "cost_micro_usd": 420,
      "timestamp": "2026-04-11T22:00:00Z"
    }
  ]
}

Request examples

Chat completions

Stream a response, continue a conversation, or set a system message.

curl -X POST https://api.demac.example/v1/chat/completions \
  -H "Authorization: Bearer <YOUR_API_KEY>" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "<model-id-from-/v1/models>",
    "messages": [{"role": "user", "content": "Explain quantum computing"}],
    "stream": true,
    "max_tokens": 1024
  }'

List models

Find model IDs and current provider coverage before sending a request.

from openai import OpenAI

client = OpenAI(base_url="https://api.demac.example/v1", api_key="<YOUR_API_KEY>")

models = client.models.list()
for model in models.data:
    print(f"{model.id}")

Need to change your example URL?

Open settings