Mistral Moderation 2411
Mistral Moderation 2411 is the first Mistral moderation API model. It classifies text into nine policy categories and can use the conversation as context for its assessment.
Deprecated API; check access details
Conversation context changes the input
The launch describes two endpoints: one for raw text and one for conversation content. In the conversation mode, the model classifies the last message using the preceding discussion. This can distinguish a quoted or discussed passage from an isolated text fragment.Mistral Moderation 2411 release announcement
Scores support an application policy
The classifier covers model-related risks such as unqualified advice and personal information as well as other harmful content categories. It returns classification information. The application still decides which categories and thresholds should block, review or permit a message.Mistral Moderation 2411 release announcement
Specifications & access
| Detail | Value and conditions |
|---|---|
| Status | Deprecated API; check access details |
| Launch date | November 2024 |
| Access | API |
| Model identifiers | mistral-moderation-2411 |
| Context window | 8,000 tokens The model catalog advertises 8k context for this version. This is not a measured recall limit. Mistral Moderation 2411 model and lifecycle record |
| Weight access | No public weight download is listed in the official model record.Mistral Moderation 2411 model and lifecycle record |
| Access and lifecycle | Hosted API deprecated on 31 March 2026.Mistral Moderation 2411 model and lifecycle record |
| Input range | Text; 8K advertised contextMistral Moderation 2411 model and lifecycle record |