API reference

AI models

Each step of the content pipeline (outlines, drafting, keyword analysis, images and more) runs on a configurable AI model. Any token can read the configuration; only admin tokens can change it, and changes apply platform-wide.

2 endpointsBase URL https://autoranko.com/api/public/v1Every request needs a bearer token

List AI models

GEThttps://autoranko.com/api/public/v1/ai/models

Shows which AI model each step of the content pipeline uses. Each entry in tasks has the model in effect (model) and its built-in default (default); kind says whether the task runs a text model (chat) or an image model (image). chat_models and image_models list the models a task can be switched to.

The configuration is platform-wide and read-only for regular accounts. It's useful for showing which model wrote your content.

Request

No parameters. Send the request with your token, nothing else.

curl https://autoranko.com/api/public/v1/ai/models \
  -H "Authorization: Bearer $AUTORANKO_TOKEN"

const res = await fetch("https://autoranko.com/api/public/v1/ai/models", {
  headers: { Authorization: `Bearer ${process.env.AUTORANKO_TOKEN}` },
});
const data = await res.json();
if (!res.ok) throw new Error(`${data.error.code}: ${data.error.message}`);

for (const task of data.tasks) {
  const note = task.model === task.default ? "" : `(default: ${task.default})`;
  console.log(task.id.padEnd(14), task.model, note);
}

import os
import requests

res = requests.get(
    "https://autoranko.com/api/public/v1/ai/models",
    headers={"Authorization": f"Bearer {os.environ['AUTORANKO_TOKEN']}"},
    timeout=30,
)
data = res.json()
if not res.ok:
    raise RuntimeError(f"{data['error']['code']}: {data['error']['message']}")

for task in data["tasks"]:
    note = "" if task["model"] == task["default"] else f"(default: {task['default']})"
    print(f"{task['id']:<14} {task['model']} {note}")

<?php
$ch = curl_init('https://autoranko.com/api/public/v1/ai/models');
curl_setopt_array($ch, [
    CURLOPT_RETURNTRANSFER => true,
    CURLOPT_HTTPHEADER => ['Authorization: Bearer ' . getenv('AUTORANKO_TOKEN')],
]);
$data = json_decode(curl_exec($ch), true);
if (isset($data['error'])) throw new RuntimeException($data['error']['message']);

foreach ($data['tasks'] as $task) {
    $note = $task['model'] === $task['default'] ? '' : " (default: {$task['default']})";
    printf("%-14s %s%s\n", $task['id'], $task['model'], $note);
}

Response

200OKThe pipeline tasks with their models, and the model catalog.

Response bodyapplication/json

tasksarray of objects
Pipeline tasks and the model each one uses.
tasks items(6)AiTask
- idstring
  Task ID. This is what PUT /ai/models takes as task.
  One ofdraftoutlinekeywordssite_analysispersonascontent_planenrichmentrewritelinkingseo_metaimage
- labelstring
  Human-readable name.
- kindstring
  chat tasks run a model from chat_models; the image task runs one from image_models.
  One ofchatimage
- descriptionstring
  What the task does in the pipeline.
- modelstring
  The model the task currently uses.
- defaultstring
  The task's built-in default model.
chat_modelsarray of objects
Models available to chat tasks.
chat_models items(2)AiModel
- idstring
  Model ID. Pass it as model to PUT /ai/models.
- labelstring
  Human-readable name and positioning.
image_modelsarray of objects
Models available to the image task.
image_models items(2)AiModel
- idstring
  Model ID. Pass it as model to PUT /ai/models.
- labelstring
  Human-readable name and positioning.

Example

Current configuration

GET/ai/models

Response · 200 OK

{
  "tasks": [
    {
      "id": "draft",
      "label": "Article drafting",
      "kind": "chat",
      "description": "Writes the full article HTML from the outline",
      "model": "gpt-6-astra",
      "default": "claude-opus-5-5"
    },
    {
      "id": "outline",
      "label": "Article outlines",
      "kind": "chat",
      "description": "Title, meta, sections, FAQs, image plan",
      "model": "gpt-6-luna",
      "default": "claude-sonnet-5-5"
    },
    {
      "id": "keywords",
      "label": "Keyword analysis",
      "kind": "chat",
      "description": "Keyword ideas, volume/difficulty estimates",
      "model": "gpt-6-luna",
      "default": "claude-haiku-5-5"
    },
    {
      "id": "site_analysis",
      "label": "Site analysis",
      "kind": "chat",
      "description": "Business profile & language detection at onboarding",
      "model": "gpt-6-luna",
      "default": "gpt-6-luna"
    },
    {
      "id": "personas",
      "label": "Buyer personas",
      "kind": "chat",
      "description": "Audience personas that steer tone & examples",
      "model": "gpt-6-luna",
      "default": "claude-sonnet-5-5"
    },
    {
      "id": "content_plan",
      "label": "Content planning",
      "kind": "chat",
      "description": "30-day topic calendar generation",
      "model": "gpt-6-luna",
      "default": "claude-sonnet-5-5"
    },
    {
      "id": "enrichment",
      "label": "Content enrichment",
      "kind": "chat",
      "description": "Improvement suggestions for existing posts",
      "model": "gpt-6-luna",
      "default": "claude-sonnet-5-5"
    },
    {
      "id": "rewrite",
      "label": "Full rewrites",
      "kind": "chat",
      "description": "Rewrites underperforming articles",
      "model": "gpt-6-astra",
      "default": "claude-opus-5-5"
    },
    {
      "id": "linking",
      "label": "Internal linking",
      "kind": "chat",
      "description": "Anchor placement across existing content",
      "model": "gpt-6-luna",
      "default": "claude-haiku-5-5"
    },
    {
      "id": "seo_meta",
      "label": "SEO metadata",
      "kind": "chat",
      "description": "Meta descriptions, focus keywords, categorization, E-E-A-T",
      "model": "gpt-6-luna",
      "default": "claude-haiku-5-5"
    },
    {
      "id": "image",
      "label": "Image generation",
      "kind": "image",
      "description": "Hero & in-article images",
      "model": "nano-banana-2",
      "default": "nano-banana-2"
    }
  ],
  "chat_models": [
    {
      "id": "claude-opus-5-5",
      "label": "Claude Opus 5.5 — highest quality"
    },
    {
      "id": "claude-fable-5",
      "label": "Claude Fable 5"
    },
    {
      "id": "claude-sonnet-5-5",
      "label": "Claude Sonnet 5.5 — balanced"
    },
    {
      "id": "claude-haiku-5-5",
      "label": "Claude Haiku 5.5 — fast & cheap"
    },
    {
      "id": "claude-sonnet-5",
      "label": "Claude Sonnet 5"
    },
    {
      "id": "gpt-6-astra",
      "label": "GPT-6 Astra — most capable GPT"
    },
    {
      "id": "gpt-6-luna",
      "label": "GPT-6 Luna"
    },
    {
      "id": "gpt-6-sol",
      "label": "GPT-6 Sol"
    },
    {
      "id": "gpt-5.6-luna",
      "label": "GPT-5.6 Luna"
    }
  ],
  "image_models": [
    {
      "id": "nano-banana-2",
      "label": "Nano Banana 2 — 2K, best default"
    },
    {
      "id": "gpt-image-2",
      "label": "GPT Image 2"
    },
    {
      "id": "seedream-3",
      "label": "Seedream 3.0"
    },
    {
      "id": "z-image",
      "label": "Z-Image — fastest"
    }
  ]
}

Errors

Error	When
401missing_token	The `Authorization: Bearer …` header is missing.
401invalid_token	The token is malformed, unknown or revoked.
429rate_limited	The token made more than 120 requests in the last 60 seconds.

Set a task's AI model

PUThttps://autoranko.com/api/public/v1/ai/modelsAdmin token

Changes the AI model one pipeline task uses. Admin only: the token must belong to an Autoranko administrator account, and tokens from regular accounts get 403 forbidden. The setting is platform-wide and takes effect from the next AI call, for every site.

The model must fit the task's kind: pick from chat_models for chat tasks and from image_models for the image task (see GET /ai/models).

The change applies to every site on the platform, immediately.

Request

Body parametersapplication/json

taskstringRequired
The pipeline task to change: an id from tasks in GET /ai/models.
One ofdraftoutlinekeywordssite_analysispersonascontent_planenrichmentrewritelinkingseo_metaimage
Example
outline
modelstringRequired
The model to use: an id from chat_models for chat tasks, or from image_models for the image task.
Example
claude-sonnet-5-5

curl -X PUT https://autoranko.com/api/public/v1/ai/models \
  -H "Authorization: Bearer $AUTORANKO_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "task": "outline",
    "model": "claude-sonnet-5-5"
  }'

const res = await fetch("https://autoranko.com/api/public/v1/ai/models", {
  method: "PUT",
  headers: {
    Authorization: `Bearer ${process.env.AUTORANKO_TOKEN}`,
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    task: "outline",
    model: "claude-sonnet-5-5",
  }),
});
const data = await res.json();
if (!res.ok) throw new Error(`${data.error.code}: ${data.error.message}`);

console.log(`${data.task} now uses ${data.model}`);

import os
import requests

res = requests.put(
    "https://autoranko.com/api/public/v1/ai/models",
    headers={"Authorization": f"Bearer {os.environ['AUTORANKO_TOKEN']}"},
    json={
        "task": "outline",
        "model": "claude-sonnet-5-5",
    },
    timeout=30,
)
data = res.json()
if not res.ok:
    raise RuntimeError(f"{data['error']['code']}: {data['error']['message']}")

print(f"{data['task']} now uses {data['model']}")

<?php
$ch = curl_init('https://autoranko.com/api/public/v1/ai/models');
curl_setopt_array($ch, [
    CURLOPT_RETURNTRANSFER => true,
    CURLOPT_CUSTOMREQUEST => 'PUT',
    CURLOPT_HTTPHEADER => [
        'Authorization: Bearer ' . getenv('AUTORANKO_TOKEN'),
        'Content-Type: application/json',
    ],
    CURLOPT_POSTFIELDS => json_encode([
        'task' => 'outline',
        'model' => 'claude-sonnet-5-5',
    ]),
]);
$data = json_decode(curl_exec($ch), true);
if (isset($data['error'])) throw new RuntimeException($data['error']['message']);

echo "{$data['task']} now uses {$data['model']}\n";

Response

200OKThe model was saved.

Response bodyapplication/json

okboolean
Always true.
taskstring
The task that changed.
One ofdraftoutlinekeywordssite_analysispersonascontent_planenrichmentrewritelinkingseo_metaimage
modelstring
The model it now uses.

Examples

PUT/ai/models

Request body

{
  "task": "outline",
  "model": "claude-sonnet-5-5"
}

Response · 200 OK

{
  "ok": true,
  "task": "outline",
  "model": "claude-sonnet-5-5"
}

PUT/ai/models

Request body

{
  "task": "outline",
  "model": "claude-sonnet-5-5"
}

Response · 403 Forbidden

{
  "error": {
    "code": "forbidden",
    "message": "Only admin tokens can change platform AI models."
  }
}

PUT/ai/models

Request body

{
  "task": "image",
  "model": "claude-sonnet-5-5"
}

Response · 400 Bad Request

{
  "error": {
    "code": "invalid_body",
    "message": "Unknown model for this task kind."
  }
}

Errors

Error	When
400invalid_body	The body isn't valid JSON (`Invalid JSON body.`).
400invalid_body	`task` isn't one of the task IDs (`Unknown task.`).
400invalid_body	`model` isn't available for the task's kind (`Unknown model for this task kind.`).
401missing_token	The `Authorization: Bearer …` header is missing.
401invalid_token	The token is malformed, unknown or revoked.
403forbidden	The token doesn't belong to an administrator account.
429rate_limited	The token made more than 120 requests in the last 60 seconds.

API version v1. Every path on this page is relative to https://autoranko.com/api/public/v1.

List AI models#

Request

Response

Example

Errors

Set a task's AI model#

Request

Response

Examples

Errors

List AI models

Set a task's AI model