API Console

DICOMPUTE API — OpenAI-compatible. Swap your base URL, keep your existing code. Every request is end-to-end encrypted and processed on hardware-attested Apple Silicon.

Endpoint Reference

Expand each endpoint to see request/response format and notes.

Base URL

https://api.dicompute.ai/v1

All endpoints are relative to this base URL. Provider attestation and pricing endpoints are publicly accessible without authentication.

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.

Quick Start

Install the OpenAI SDK or Vercel AI SDK. The DICOMPUTE API is fully OpenAI-compatible — just change the base URL.

# No installation needed
export DARKBLOOM_API_KEY="<YOUR_API_KEY>"
export DARKBLOOM_BASE_URL="https://api.dicompute.ai/v1"

Available Models

Model ID	Type	Architecture
<model-id-from-/v1/models>	text	Returned by /v1/models

Model availability depends on online providers. Check /v1/models for real-time availability.

Chat Completions

Stream chat completions with any supported model. Supports system messages, multi-turn conversations, and thinking/reasoning output.

curl -X POST https://api.dicompute.ai/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

Check available models, provider counts, and attestation status.

from openai import OpenAI

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

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