Netflix unveils GenRec, an LLM-native recommendation ranker replacing traditional RecSys

GenRec: Towards LLM-Native Recommendation at Netflix

Netflix built GenRec, an LLM that directly ranks recommendations instead of using a traditional RecSys pipeline. It first continues training a foundation LLM on Netflix data, then fine-tunes it to take user watch history and candidate titles as natural-language input and output ranking scores. In online A/B tests, GenRec lifted watch time by 0.8% over the production baseline, with offline NDCG up 2.3%. Inference latency is kept under 11 ms. The post does not disclose the base model name or parameter count.

Why it matters: Netflix replacing its recsys pipeline with an LLM is a concrete engineering move from a major tech company—real substance. But no open source, no public benchmarks, so impact stays within the recsys community. Clears the featured bar but not higher.