POST/shorten2 creditsMidjourney only

Shorten

Analyze a long Midjourney prompt and identify which tokens carry the most weight. Returns multiple shortened variants that retain the most impactful elements, along with a full markdown analysis showing token importance scores. Use this to clean up over-specified prompts, understand what's actually driving the output, and build leaner prompt templates. Midjourney only.

Request Parameters

ParameterTypeRequiredDescription
promptstringRequiredThe prompt to analyze and shorten.
webhook_urlstringOptionalURL where we POST the completed task result.
webhook_secretstringOptionalSent as x-webhook-secret header in webhook delivery.

Example Request

cURL
curl -X POST https://api.journeyapi.com/api/v1/shorten \
  -H "Authorization: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "a beautiful stunning gorgeous amazing incredible photograph of a sunset over mountains with golden light",
    "webhook_url": "https://your-server.com/webhook"
  }'

Immediate Response

Returns immediately with a task ID. Use /fetch or webhooks to get the result.

200 OK
{
  "task_id": "gg1j9f11-j30m-52o5-l827-557766551111",
  "task_type": "shorten",
  "content": [
    "1️⃣ photograph of sunset over mountains, golden light",
    "2️⃣ stunning photograph, sunset mountains, golden light",
    "3️⃣ sunset over mountains, golden light photograph"
  ],
  "full_content": "**Important tokens:** photograph, sunset, mountains, golden light\n**Low impact:** beautiful, stunning, gorgeous, amazing, incredible"
}

Response Fields

FieldTypeDescription
task_idstringUUID of the shorten task.
task_typestringAlways "shorten".
contentstring[]Array of shortened prompt variants.
full_contentstringFull markdown-formatted analysis with token importance.

Tips

  • 1Use shorten to optimize verbose prompts. Midjourney often ignores filler adjectives.
  • 2The full_content field shows which tokens have high vs low impact.

Quirks & Gotchas

Content items are prefixed with emoji numbers.

Expert Tips & Best Practices

Identifying high-weight tokens in your prompt

The full_content field returns a markdown analysis showing which prompt tokens Midjourney weighted most heavily. Tokens near the front and tokens that are visually dominant in Midjourney's training data score higher. Use this to diagnose why your output doesn't match your intent: if a low-priority word is overriding your main subject, move it later in the prompt or remove it.

Cleaning up over-specified prompts

Long prompts with many adjectives often produce muddier results than focused prompts. Run your prompt through /shorten and compare the shortened variants against the original — the shorter versions often outperform the original because they remove conflicting or redundant tokens that dilute Midjourney's focus. Keep the 5–7 highest-weight tokens and drop the rest.

Related Endpoints