Musubi launches PolicyLM-1.7B for real-time content moderation

The model is meant to let platform teams update moderation rules without retraining, so policy changes can be reflected faster.

Musubi has released PolicyLM-1.7B, an open-weights decision model built for real-time content moderation, and the company says it can apply written policies to messages in under 50 milliseconds. The model is meant to give platform operators a scalable way to classify content without retraining every time rules change, which could make moderation policy updates much easier to iterate. Musubi co-founder and chief AI officer Filip Jankovic says the company is aiming at proactive labeling for fast-growing platforms. Jankovic also says his work on decision models started before Jev, tracing back to a 2024 project called GLiNER.