MagnetAPI.org Customer API Documentation

Bash: set API_KEY to a System Token or User API Key from this site's dashboard before running these commands.

curl --fail-with-body https://api.magnetapi.org/v1/models \
  -H "Authorization: Bearer $API_KEY"

Read the assistant reply from choices[0].message.content. Inspect tool_calls if the reply asks to use a function.

curl --fail-with-body --max-time 180 https://api.magnetapi.org/v1/chat/completions \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"power-ai","max_tokens":64,"messages":[{"role":"user","content":"Reply exactly OK."}]}'

Read text from output items containing output_text blocks. Response IDs and usage totals vary by request.

curl --fail-with-body --max-time 180 https://api.magnetapi.org/v1/responses \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"power-ai","input":"Reply exactly OK.","max_output_tokens":64}'

Use /v1/messages and include the anthropic-version header. Read text blocks from content. No dashboard model change is needed when the request explicitly names power-ai.

curl --fail-with-body --max-time 180 https://api.magnetapi.org/v1/messages \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -H "anthropic-version: 2023-06-01" \
  -d '{"model":"power-ai","max_tokens":64,"messages":[{"role":"user","content":"Reply exactly OK."}]}'

This Responses example keeps the connection open for server-sent events. Handle text deltas, response.completed and error events; HTTP 200 alone does not prove a successful generation. Messages streams use content_block_delta and message_stop instead.

curl --fail-with-body --max-time 180 -N https://api.magnetapi.org/v1/responses \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"power-ai","input":"Reply exactly OK.","max_output_tokens":64,"stream":true}'

Your application must execute get_project_name locally. Return a role=tool message with the matching tool_call_id after the complete assistant tool_calls message, then submit the conversation again. Do not omit a tool result or execute an unapproved function. Tools and images remain subject to the active destination's capabilities.

curl --fail-with-body --max-time 180 https://api.magnetapi.org/v1/chat/completions \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"power-ai","messages":[{"role":"user","content":"Use get_project_name to read the current project name."}],"tools":[{"type":"function","function":{"name":"get_project_name","description":"Read the name of the current project.","parameters":{"type":"object","properties":{},"additionalProperties":false}}}],"tool_choice":{"type":"function","function":{"name":"get_project_name"}},"max_tokens":256}'

Python 3, standard library only. Set API_KEY in your environment. This sends one request and prints the answer and support request ID.

import json
import os
from urllib.request import Request, urlopen

payload = {'model': 'power-ai', 'max_tokens': 64, 'messages': [{'role': 'user', 'content': 'Reply exactly OK.'}]}
request = Request(
    "https://api.magnetapi.org/v1/chat/completions",
    data=json.dumps(payload).encode("utf-8"),
    headers={"Authorization": "Bearer " + os.environ["API_KEY"],
             "Content-Type": "application/json"},
    method="POST",
)
with urlopen(request, timeout=180) as response:
    result = json.load(response)
    print(result["choices"][0]["message"]["content"])
    print("Request ID:", response.headers.get("x-request-id", "not supplied"))

Run with Node.js 18 or later. Keep the key on your server, never in public browser JavaScript. Set API_KEY before running the script.

async function main() {
  if (!process.env.API_KEY) throw new Error("Set API_KEY first");
  const response = await fetch("https://api.magnetapi.org/v1/chat/completions", {
    method: "POST",
    headers: {Authorization: `Bearer ${process.env.API_KEY}`, "Content-Type": "application/json"},
    body: JSON.stringify({"model": "power-ai", "max_tokens": 64, "messages": [{"role": "user", "content": "Reply exactly OK."}]}),
    signal: AbortSignal.timeout(180000),
  });
  const requestId = response.headers.get("x-request-id");
  if (!response.ok) throw new Error(`HTTP ${response.status}; request ID ${requestId}`);
  const result = await response.json();
  console.log(result.choices[0].message.content);
  console.log("Request ID:", requestId);
}
main().catch(error => { console.error(error.message); process.exitCode = 1; });

Set the API_KEY environment variable to your own key, then paste this into PowerShell. This example does not overwrite any CLI configuration.

$ErrorActionPreference = 'Stop'
if (-not $env:API_KEY) { throw 'Set API_KEY first' }
$body = @{
  model = 'power-ai'
  max_tokens = 64
  messages = @(@{ role = 'user'; content = 'Reply exactly OK.' })
} | ConvertTo-Json -Depth 10
$result = Invoke-RestMethod -Method Post -Uri 'https://api.magnetapi.org/v1/chat/completions' -TimeoutSec 180 -Headers @{ Authorization = "Bearer $env:API_KEY" } -ContentType 'application/json' -Body $body
$result.choices[0].message.content