Get your API key

Sign up and open Dashboard → API Keys. Copy the secret when shown — you can reveal saved keys later.

Make your first request

curl https://api.hadexs.com/v1/chat/completions \
  -H "Authorization: Bearer hadexs_sk_..." \
  -H "Content-Type: application/json" \
  -d '{
    "model": "hades-1.0",
    "messages": [{"role": "user", "content": "Explain quantum computing in simple terms"}]
  }'

Use the Python SDK

from openai import OpenAI

client = OpenAI(
    api_key="hadexs_sk_...",
    base_url="https://api.hadexs.com/v1",
)

completion = client.chat.completions.create(
    model="hades-1.0",
    messages=[{"role": "user", "content": "Hello!"}],
)
print(completion.choices[0].message.content)

Use with Node.js

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: "hadexs_sk_...",
  baseURL: "https://api.hadexs.com/v1",
});

const response = await client.chat.completions.create({
  model: "hades-1.0",
  messages: [{ role: "user", content: "Hello!" }],
});
console.log(response.choices[0].message.content);

API Keys

Create keys in the dashboard. Each key begins with hadexs_sk_.

Request Headers

Authorization: Bearer hadexs_sk_...

Security Best Practices

• Never expose API keys in client-side browser code
• Store keys in environment variables, not source code
• Rotate keys immediately if compromised
• Create separate keys per application or environment
• Delete unused keys from your dashboard

Create a key

Open Dashboard → API Keys and click Create API Key. Name keys by environment (Production, Staging, Local).

Reveal & rotate

Use the eye icon to reveal a saved key. Delete and recreate to rotate.

Dashboard

Manage keys, usage, and billing from the developer dashboard.

Limits by Tier

Rate Limit Headers

x-ratelimit-limit-requests: 30
x-ratelimit-limit-tokens: 14400
x-ratelimit-remaining-requests: 28
x-ratelimit-remaining-tokens: 13122
x-ratelimit-reset-requests: 2s
x-ratelimit-reset-tokens: 1s

Handling 429 Errors

Retry with exponential backoff or upgrade your plan for higher limits.

Request Body

Example Request

curl https://api.hadexs.com/v1/chat/completions \
  -H "Authorization: Bearer hadexs_sk_..." \
  -H "Content-Type: application/json" \
  -d '{
    "model": "hades-1.0",
    "messages": [
      {"role": "system", "content": "You are a helpful assistant."},
      {"role": "user", "content": "What is 2 + 2?"}
    ],
    "temperature": 0.7,
    "max_tokens": 1024
  }'

from openai import OpenAI
client = OpenAI(api_key="hadexs_sk_...", base_url="https://api.hadexs.com/v1")
print(client.chat.completions.create(
  model="hades-1.0",
  messages=[{"role": "user", "content": "What is 2 + 2?"}],
).choices[0].message.content)

import OpenAI from "openai";
const client = new OpenAI({ apiKey: "hadexs_sk_...", baseURL: "https://api.hadexs.com/v1" });
const res = await client.chat.completions.create({
  model: "hades-1.0",
  messages: [{ role: "user", content: "What is 2 + 2?" }],
});
console.log(res.choices[0].message.content);

Example Response

{
  "id": "chatcmpl-abc123",
  "object": "chat.completion",
  "model": "hades-1.0",
  "choices": [{
    "index": 0,
    "message": { "role": "assistant", "content": "2 + 2 equals 4." },
    "finish_reason": "stop"
  }],
  "usage": { "prompt_tokens": 24, "completion_tokens": 9, "total_tokens": 33 }
}

Language Models

Audio Models

List Models via API

curl https://api.hadexs.com/v1/models \
  -H "Authorization: Bearer hadexs_sk_..."

Enable streaming

Set stream: true on chat completion requests.

stream = client.chat.completions.create(
  model="hades-1.0",
  messages=[{"role": "user", "content": "Tell me a story"}],
  stream=True,
)
for chunk in stream:
  if chunk.choices[0].delta.content:
    print(chunk.choices[0].delta.content, end="", flush=True)

Stream Format

Each SSE line is a JSON object with choices[0].delta.content fragments until [DONE].

Endpoint

Request

Pass prompt, model (hades-image or tiered ids), and optional size.

Image models

hades-1.5 and hades-2.0 image tiers map to your plan. See Pricing for daily image allowances.

Tools parameter

Define JSON-schema functions the model may invoke; handle tool calls in your app loop.

Example

See OpenAI tool calling docs — Hades accepts the same request shape at /v1/chat/completions.

HTTP status codes

• 401 — invalid or missing API key
• 429 — rate limit exceeded
• 500 — server error (retry with backoff)

Error shape

{ "error": { "message": "...", "type": "invalid_request_error", "code": "..." } }

Keys & security

Use server-side proxies; never ship hadexs_sk_ keys to browsers.

Retries

Retry idempotent requests on 429/5xx with jittered exponential backoff.

Production

Separate keys per service, monitor usage in the dashboard, and set wallet alerts before launch.

Guided walkthrough

The terminal is always the source of truth. Sign in with the Hadexs account that holds your Pro (or higher) subscription, then run agents against the folder you are in.

# 1. Install the CLI
$ npm install --global https://hadexs.com/downloads/cli.tgz

# 2. Sign in with your Hades account
$ hzunos login you@email.com

# 3. Run a task from your project folder
$ hzunos run "Research the top 10 competitors and write a report"

→ connected to Hades · local execution enabled
✓ report saved to ./competitor-report.md

Step by step

The frames below show the normal path through the CLI.

$ npm install --global https://hadexs.com/downloads/cli.tgz
added zunos@0.1.0
$ hzunos --help
✓ hzunos is ready

$ hzunos login you@email.com
→ Checking entitlement…
✓ Logged in as you@email.com — plan pro
$ hzunos whoami
  email   you@email.com
  plan    pro
  paid    true

$ hzunos agents
  chief     Orchestrator
  coder     Engineering
  shell     Terminal operator
  research  Analyst
$ hzunos run --agent research "Summarize this repo"

$ hzunos shell --agent coder
chief › Explain package.json scripts
✓ reply streamed in this terminal
coder › exit

$ hzunos run "Write the competitor report to research/report.md"
✓ info — sources collected
✓ write_file — research/report.md
✓ task complete — 2 actions
  Saved inside the current workspace.

× Paid Zunos plan required (you@email.com)
→ hzunos status
  plan    free
  paid    false
· Open https://hadexs.com/pricing to upgrade

Commands

Everything the binary understands. zunos works anywhere you type hzunos.

Agents

Plans & access

• Install once globally; re-run the npm install URL to update.
• Prefer the project root as your cwd so relative paths stay predictable.
• Pair with the IDE extension for Command Palette launches.
• Desktop Mac app is separate — download from /download (also paid).

Recent releases

API v1 chat, streaming, dashboard billing, and docs refresh — October 2026.

X (Twitter)

@DjKenpz6u — announcements and build updates.

Instagram

@hadessai — product and company posts.

Contact

Email support@hadexs.com or zunosagents@gmail.com, or visit the contact page.

Uptime

Live checks at hadexs.com/status. Incident updates post there first.

Contact

Email support@hadexs.com or zunosagents@gmail.com. For Zunos agent workflows, see Zunos.