Which open-source model would you recommend for generating video recommendations in an OTT platform?

TensorFlow Recommenders (TFRS) with a two-tower retrieval model is the recommended open-source starting point for video recommendation in an OTT platform, according to a technical response. The response suggests using TFRS's basic retrieval tutorial, and for sequential watch histories, BERT4Rec or SASRec via RecBole, with sentence-transformers/all-MiniLM-L6-v2 for cold-start content embeddings. The recommended progression is to start with TFRS two-tower retrieval plus a ranker, and only adopt BERT4Rec if sequential data improves offline metrics.

I would not start with a general-purpose LLM for this. Video recommendation is mainly a retrieval-and-ranking problem, and the best model depends on what interaction data you have. A practical open-source starting point is TensorFlow Recommenders TFRS with a two-tower retrieval model: The official movie-retrieval tutorial is close to this use case and includes training, evaluation, and export: https://www.tensorflow.org/recommenders/examples/basic retrieval https://www.tensorflow.org/recommenders/examples/basic retrieval If you have ordered watch histories, compare sequential models such as BERT4Rec or SASRec through RecBole. BERT4Rec is specifically designed to model sequences of user interactions: BERT4Rec — RecBole 1.2.1 documentation https://recbole.io/docs/user guide/model/sequential/bert4rec.html For a cold-start baseline, you can embed titles, descriptions, genres, and tags with a text encoder such as sentence-transformers/all-MiniLM-L6-v2 and recommend semantically similar videos. Its model card lists semantic search as an intended use: sentence-transformers/all-MiniLM-L6-v2 · Hugging Face https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2 . However, that is an item-content encoder, not a complete personalized recommender. My suggested progression would be: So, if I had to choose one free starting stack: TFRS two-tower retrieval plus a ranker , with content embeddings for cold start. I would choose BERT4Rec only after confirming that sequential watch history improves the offline metrics.