A Sightengine alternative with credits for each kind of content.

What Sightengine says about price and plans

From Sightengine’s pricing page, FAQ and documentation, as read on 17 October 2026. Detectivision is not affiliated with Sightengine.

The prices above include the tiers of the pricing page’s plan sliders. Sightengine can change them, so check its page before you decide.

How Sightengine counts operations

From Sightengine’s FAQ, as read on 17 October 2026. Sightengine adds that the counts for visual models “may not apply to accounts on legacy models (from before 2025)”.

What Sightengine checks, and what comes back

From Sightengine’s documentation and FAQ, as read on 17 October 2026.

When Sightengine fits

Sightengine’s own pages show what it offers. These are the cases where that list decides it.

• You want a free plan with no end date

  Free gives 2,000 operations a month, at most 500 a day, at 1 request a second, and Sightengine says there is no time limit. It includes basic content moderation and AI detection.

• You pick from a large library of image models

  Sightengine lists models for nudity in 29 classes, violence, gore, weapons, drugs, alcohol, tobacco, gambling, money, hate signs, destruction, military and religious content, and you choose the ones you need for each request.

• You need AI-generated media or identity checks

  Sightengine lists AI-generated image and video detection and deepfake detection on Starter, and face liveness, age estimation and AI music detection on Pro.

• You write text rules in many languages

  Its rule-based text moderation covers 18 languages, returns the flagged words with an intensity for profanity, and takes lists of your own.

• You keep lists of images and videos

  Pro allows 10 custom lists. A list for image search or near-duplicate detection holds up to 1,000,000 images or 1,000 videos.

• You need enterprise terms

  Enterprise plans can choose where data is processed, and come with SAML single sign-on, service level agreements and a dedicated support engineer.

When Detectivision fits

What Detectivision documents, point by point. The tables show Sightengine’s side of each.

• Text checks in ten languages →

  Detectivision documents its text checks for English, Spanish, Portuguese, French, German, Italian, Turkish, Arabic, Russian and Hindi, with no language to set in the request.

• An image costs the same, whichever flags →

  An image costs 1 credit when the result comes later by webhook and 2 when you wait for the answer, whether it is checked for one flag or for all of them.

• Speech checks in ten languages →

  Detectivision’s audio moderation understands the same ten languages as its text moderation. In capture mode, the default, each finding comes with its start and end in seconds.

• Video, audio and live streams in one credit system →

  Every plan has video credits, 300 on Basic. Pro adds 2,000 audio and 1,000 live stream credits. A live session can be set up to 30 frames a second and keeps flagged moments with their picture.

• A review screen with your own actions →

  A moderator opens a request in the dashboard’s Logs and takes one of your actions, such as Approve or Ban. Your system gets each one as a moderation.action webhook and carries it out.

Sightengine and Detectivision, row by row

Sightengine from its pricing page, FAQ and docs, as read on 17 October 2026. Detectivision from its docs and pricing page.

Units differ: Sightengine counts operations, Detectivision credits. The examples count images and texts, not dollars: Starter is $29 a month and Basic $49. Work out a busy month on both price pages before you choose.

Where Sightengine’s models go

Sightengine’s models and classes as its docs list them on 17 October 2026, next to the Detectivision flags that cover the same ground. GET /api/v1/flags lists the flags of your account. A flag keeps its key.

The table lists the models and classes that have a matching flag. Sightengine’s model pages list more; compare them with the list your account gets from GET /api/v1/flags before you move.

An image check, both ways

curl -X GET -G 'https://api.sightengine.com/1.0/check.json' \
  -d 'models=nudity-2.1,violence,offensive-2.0' \
  -d 'api_user={api_user}&api_secret={api_secret}' \
  --data-urlencode 'url=https://example.com/photo.jpg'

curl https://api.detectivision.ai/api/v1/moderation \
  -H "Api-Key: $DETECTIVISION_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "type": "image_moderation",
    "input": "https://example.com/photo.jpg",
    "mode": "detect",
    "flags": ["sexual", "violence", "hate"]
  }'

Sightengine takes the models in models and an API user and secret pair in the request. Detectivision takes the flags in flags and its key in the Api-Key header. The three Sightengine models are three operations. The Detectivision check as written is answered at once and costs 2 credits; with "async": true it costs 1.

A text check, both ways

curl -X POST 'https://api.sightengine.com/1.0/text/check.json' \
  -F 'text=You are such an idiot, just shut up.' \
  -F 'lang=en' \
  -F 'models=general,self-harm' \
  -F 'mode=ml' \
  -F 'api_user={api_user}' \
  -F 'api_secret={api_secret}'

curl https://api.detectivision.ai/api/v1/moderation \
  -H "Api-Key: $DETECTIVISION_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "type": "text_moderation",
    "input": "You are such an idiot, just shut up.",
    "mode": "detect",
    "flags": ["insult", "hate", "sexual", "self_harm"]
  }'

Sightengine asks for the language in lang and the models in models. Detectivision reads the text in input with no language to set, and checks the flags you name; without flags, every text flag of your account.

What comes back

The fields Sightengine’s docs describe, next to the fields of a Detectivision answer (docs: Responses).

At once, or later by webhook

{
  "id": 4120,
  "type": "image_moderation",
  "mode": "detect",
  "status": "completed",
  "flagged": true,
  "flags": [{ "flag": "violence", "confidence": 0.87 }],
  "checked": ["hate", "scam", "sexual", "violence"],
  "unchecked": [],
  "tookMs": 1240,
  "usage": { "charged": 2, "remaining": 5126 }
}

curl https://api.detectivision.ai/api/v1/moderation \
  -H "Api-Key: $DETECTIVISION_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "type": "image_moderation",
    "input": "https://example.com/photo.jpg",
    "mode": "detect",
    "async": true,
    "metadata": { "listingId": "l_7731" }
  }'

Sent async, the same check costs half: the API answers 202 with the request’s id, and the result comes by webhook with your metadata.

Moving from Sightengine in five steps

1. List the models and rules you use

   01

   Write down the models in your requests, the classes you act on, and the rules of any workflow.

2. Map them to flags →

   02

   Use the table above, then ask GET /api/v1/flags for the flags your account has. A flag keeps its key.

3. Recount your volume in credits →

   03

   A text costs 1 credit per started 2,500 characters and an image 1 credit by webhook, whichever flags it checks. An answer given at once costs twice that. Count your busiest month, not an average one.

4. Set up your own actions →

   04

   Act on flagged and the flags raised. Define actions such as Approve or Ban in the dashboard’s Logs, and your system gets each one as a moderation.action webhook.

5. Send the same content to both

   05

   Run a sample of your own messages and images through Sightengine and Detectivision, as separate calls, and compare the answers before you switch.

• Content moderation APIs compared →

  What each API checks, its languages and its published prices.

• Free content moderation APIs →

  What each vendor gives for free, and its limits.

• Hive alternative →

  Its prices, its self-serve limit and its contracts, set against credits.

• OpenAI Moderation API →

  Its limits, and when a paid API is worth it.

• Amazon Rekognition alternative →

  Rekognition’s moderation labels, mapped one by one.

• Video moderation →

  Each finding with its second and its box, a censored copy and tracking.