Unification of Capabilities

The interface of the /embeddings endpoint has been simplified by merging five separate models into a single model. This new model performs better across various text search, sentence similarity, and code search benchmarks.

Longer Context

The new model has an increased context length of four times, from 2048 to 8192. This makes it more convenient to work with long documents.

Smaller Embedding Size

The new embeddings are smaller, with only 1536 dimensions. They are one-eighth the size of davinci-001 embeddings, which makes them more cost-effective for vector database operations.

Reduced Price

The price of the new embedding models has been reduced by 90% compared to the old models of the same size. Despite the lower price, the new model achieves better or similar performance as the old Davinci models at a 99.8% lower cost.

Overall, the new embedding model is a powerful tool for natural language processing and code tasks. It offers enhanced capabilities and affordability. It will be exciting to see how customers leverage it to create even more advanced applications in their respective fields.