MongoDB believes that improved data retrieval, rather than simply larger AI models, is crucial for building trustworthy enterprise AI systems. As agentic systems and Retrieval-Augmented Generation (RAG) gain traction in production environments, the database provider has identified retrieval quality as a significant bottleneck that can negatively impact accuracy, cost-efficiency, and user confidence, even when the underlying AI models are robust.

To address this challenge, MongoDB recently launched four new versions of its embeddings and reranking models, collectively known as Voyage 4. These models are designed to enhance the accuracy and efficiency of data retrieval in AI applications. The Voyage 4 family includes voyage-4 embedding, a general-purpose model; voyage-4-large, considered the flagship model; voyage-4-lite, optimized for low-latency and cost-sensitive applications; and voyage-4-nano, intended for local development, testing, and on-device data retrieval. Voyage-4-nano marks MongoDB's first foray into open-weight models.

All Voyage 4 models are accessible via an API and on MongoDB's Atlas platform. According to MongoDB, these models outperform comparable models in retrieval quality.

The emphasis on retrieval quality highlights a growing understanding within the AI community that the effectiveness of AI systems hinges not only on the power of the models themselves but also on their ability to access and process relevant information efficiently. RAG systems, for example, rely on retrieving relevant documents or data snippets to augment the knowledge of a large language model (LLM) before generating a response. If the retrieval component is weak, the LLM may be fed inaccurate or incomplete information, leading to suboptimal results.

Agentic systems, which are designed to autonomously perform tasks, also depend on reliable data retrieval to make informed decisions and take appropriate actions. Poor retrieval quality in these systems can lead to errors, inefficiencies, and even potentially harmful outcomes.

MongoDB's focus on embeddings and reranking models reflects a trend toward optimizing the entire AI pipeline, from data ingestion to model deployment. By improving the accuracy and efficiency of data retrieval, companies can build more trustworthy and effective AI systems that deliver tangible business value. The availability of these models through an API and on the Atlas platform simplifies integration for developers and organizations looking to enhance their AI capabilities. The company stated that the models outperform similar models.