![{{24 2026-01-13 13:35:24.421 +0000 UTC 2026-01-13 13:35:24.421 +0000 UTC {0001-01-01 00:00:00 +0000 UTC false}} Alibaba Nlp alibaba nlp <nil> []}](/assets/images/wan-ai.png)
Gte Modernbert Base
GTE ModernBERT base is a compact embeddings and reranking model built on modernBERT that performs strongly on retrieval benchmarks like MTEB, LoCO, and COIR. It is designed for efficient semantic search, reranking, and embedding generation at moderate cost and low latency.
GTE ModernBERT base is a 149M-parameter embeddings and rerank model by Alibaba-NLP that focuses on high-quality semantic representations for retrieval tasks. It is tuned to excel on established benchmarks such as MTEB, LoCO, and COIR, offering a balance of strong retrieval performance and computational efficiency that makes it suitable for production search and recommendation systems. This model is versatile for tasks like semantic search, document reranking, and building embeddings pipelines where cost and speed matter. Available via serverless, on-demand, and dedicated deployments, it is a practical choice when you need reliable embedding quality for downstream retrieval tasks without the overhead of very large models.