Pinecone Testimonials

“We figured out that the base models you can get off the shelf (e.g. GPT-4) are smart enough to produce some really valuable outputs if you can get the right data into the prompt and you can explain it to the LLM the right way. And doing this comes down to retrieval.”

"The introduction of Pinecone serverless has led to amazing performance and efficiency improvements in our vector search capability. We will continue to push forward searching Billions of vectors with Pinecone serverless at the center.”