Platform & Safety
Moderation & safety
Detect harmful content, PII and prompt injection.
What it is
Moderation APIs classify text and images for harmful content (hate, violence, sexual content, self-harm), detect personal data, and increasingly screen prompts for jailbreaks and prompt injection.
How pricing works
Per request or per 1K characters/images; some are free with another API.
We compare everything in one unit: USD per 1K requests.
Typical uses
- User-generated content platforms
- Guardrails for LLM apps
- PII redaction
What to compare
- Categories covered
- False-positive rate
- Image support
- Latency
Watch out for
- No classifier is perfect; combine with human review for high-stakes decisions.
All 4 APIs, cheapest first
Free to browseTick 2 to 4 APIs to compare them side by side.
| API | $ / 1K requests | Image moderation | Prompt-injection detection | PII detection | Quality | Price data | Compare |
|---|---|---|---|---|---|---|---|
| OpenAI Moderation OpenAI | Free | Yes | No | No | 78 | Verified Oct 4, 2026 · aging | |
| Azure AI Content Safety Microsoft Azure AI | $0.38 | Yes | Yes | No | – | Unverified sample | |
| Amazon Bedrock Guardrails Amazon Web Services | $0.75 | Yes | Yes | Yes | – | Unverified sample | |
| Lakera Guard Lakera Guard | $1.00 | No | Yes | Yes | 84 | Unverified sample |