A great option for Vector databases
What do you like best about the product?
The ease of use to get integrated with Pinecone was pretty incredible. We were up and running with a vector database in no time.
What do you dislike about the product?
At first, the UI lacked some features that seemed like a must, but they've added a lot of what we were looking for and seem to be actively developing it.
What problems is the product solving and how is that benefiting you?
To perform semantic search on our documents.
I really like the product and satisfied from the ease-of-use and performance
What do you like best about the product?
I like the ease-of-use. Super easy to build index, populate with data and test it.
What do you dislike about the product?
Some security-related features are missing.
We need VPC peering in GCP, in order to unlock deals with companies that require this feature.
Also, Serverless in GCP is missing.
What problems is the product solving and how is that benefiting you?
Vector DB for multi-tenant system.
A reliable cloud solution for building an ERP dashboard
What is our primary use case?
We've been building an ERP dashboard using generative UI. We needed a vector database to retrieve and implement augmentation, so we opted to use Pinecone.
We chose Pinecone because it covers most of the use cases. Also, Pinecone is stable and reliable.
How has it helped my organization?
We are using Pinecone for retrieval. Pinecone did a really great job in marketing and perfecting its adoption. That was very helpful because we could find resources if we got stuck on a problem. The only reason we are not using Quadrant, despite its promising features and reliable performance, is its limited resources. Pinecone community has been around for a lot longer than the Quadrant community.
We chose Python so that any new feature we could add could be implemented easily. Since Python has been around for a while, plenty of options are available. Some tutorials and resources, such as blog posts, provide references for implementing new features. We haven't utilized anything specific to Pinecone that only Pinecone offers.
What is most valuable?
The tool collects data, adds it to the database, and retrieves it using its SDK.
What needs improvement?
Onboarding could be better and smoother. Navigation is difficult because most of us rely on watching tutorials on YouTube to understand how to use this software. The onboarding journey should explain more topics.
For how long have I used the solution?
We are currently using it.
What do I think about the stability of the solution?
The solution is stable. We use it for enterprise purposes. It's reliable for our use case. We haven't experienced any downtime or significant latency issues.
What do I think about the scalability of the solution?
We use the tool for a single project with 10-15 people.
Which solution did I use previously and why did I switch?
We worked with PG vector. We were using Sophos and a free directory plugin. We used it for testing and building the prototype. When I built it, we opted for more widely adopted services. We chose a database geared towards storing and retrieving active data, especially in augmenting generation.
How was the initial setup?
The initial setup is great.
What was our ROI?
The scope of the project was really small to opt for positives with the PG vector plug-in. We opted for Pinecone since it is popular, and has a better use case.
What other advice do I have?
The main issue arises when our team members join, and we must guide them, especially those unfamiliar with Pinecone. We assign them a small project to explore the software independently. This helps them overcome any hurdles and gain a deeper understanding of how to utilize Pinecone effectively. However, despite its overall positive aspects, there's room for improvement, particularly in making it more minimalistic and simplifying access to various options. Like many SaaS products, setting it up can be time-consuming. It should provide clear instructions or a step-by-step guide for undertaking small projects independently.
Real-time data retrieval is good. However, it used to drop in a while. Overall, it was reliable.
We don't require a lot of maintenance on the project. It's a small-scale project, and the scope is specific and small. There haven't been any issues. Two to Three people are enough for the solution's maintenance.
I recommend the solution and advise you to explore the documentation and tutorials. It's easy to pick up and integrate.
Overall, I rate the solution an eight out of ten.
A tool that offers its users multiple search options for retrieval purposes
What is our primary use case?
In my company, we store our industry documents in Pinecone. My company stores the PDF files in Pinecone to use for the RAG application.
What is most valuable?
The most valuable features of the solution are similarity search and maximal marginal relevance search for retrieval purposes.
What needs improvement?
The product fails to offer a serverless type of storage capacity. From an improvement perspective, the storage capacity of the tool should not be pod-based.
For how long have I used the solution?
I have been using Pinecone for years. I am an end user of the solution.
What do I think about the stability of the solution?
It is a stable solution. Stability-wise, I rate the solution an eight out of ten.
What do I think about the scalability of the solution?
Around four or five people in my company use the product.
The solution is used on a daily basis in my company.
Which solution did I use previously and why did I switch?
I don't have any previous experience with any other solutions other than Pinecone.
How was the initial setup?
The solution is deployed on an on-premises model.
The solution can be deployed in a day.
Which other solutions did I evaluate?
My company is currently evaluating Elasticsearch against Pinecone.
What other advice do I have?
My company has integrated Pinecone into our machine-learning workflow by using LangChain. My company also uses an OCR feature to detect PDF files, which we added to Pinecone.
A chatbot application is the specific AI application for which Pinecone is used in our organization since it provides us with a source of knowledge through RAG.
I am unsure if Pinecone's similar search capabilities have enhanced our data analysis since my company is still in the middle of the tool's production phase.
If I measure Pinecone's impact on our company's system performance and scalability, I would rate it at an eight on a scale of one to ten.
I rate the overall tool an eight out of ten.
One of the most convenient way for you to build a LLM-based Application
What do you like best about the product?
You can deploy pinecone very fast without caring about the backend things like docker,storage etc. with an account you can directly building your app with the offical API and python code.
What do you dislike about the product?
the price is relatively high comparing to some opensourced alternative.
What problems is the product solving and how is that benefiting you?
We are building a LLM-based Application.
Pinecone is the essential part of RAG solution.
Easy to use and powerful vector database
What do you like best about the product?
It is very easy to integrate the Pinecone API with a text generation application using LLM. Semantic search is very fast and allows more complex queries using metadata and namespace. I also like the comprehensive documentation.
What do you dislike about the product?
For organizations that need only a little more capacity than is available in a single free pod, the pricing may be more favorable.
What problems is the product solving and how is that benefiting you?
We use Pinecone as a vector database containing almost 150,000 of decisions of the Supreme Court of the Czech Republic and approximately 50 legal statutes. Pinecone serves as the backbone for the knowledge retrieval (RAG) of our legal research application.
Great dev experience
What do you like best about the product?
Easy to use
Good documentation
Easy to implement
What do you dislike about the product?
Couldn't delete an entire vector within a namespace
What problems is the product solving and how is that benefiting you?
Vector index storage provider. We store embedded indices on Pinecone.
User-friendly enterprise grade vector database
What do you like best about the product?
We started using Pinecone pretty early on. I like the light UI on top of an API-first approach. We have been using it now for millions of daily queries, and it has rarely, if ever, gone down or giving us trouble. Highly recommended!
What do you dislike about the product?
Not sure what to say here. It's been a good experience overall. If I had to say something, the pricing was tricky to groc.
What problems is the product solving and how is that benefiting you?
Fast retrieval of multi-modal search queries
fast and easy to setup vector database
What do you like best about the product?
The things I mostly like are:
- that is easy to set up by following the docs
- fast for loading and updating embeddings in the index
- easy to scale if needed
What do you dislike about the product?
- that is not open source
- I cannot query the full list of ids from an index (I needed to build a database and a script to track what products I have inside the index)
- customer support by mail takes too much time
What problems is the product solving and how is that benefiting you?
I built a deep learning model for product matching in the ecommerce industry. One of the steps for the system is to find candidates that are potential matches for the searched product. Becase of this, I needed a vector database to store the embeddings (texts and image) for the products for doing a similarity search as a first step of the product matching system.
GWI on Pinecone
What do you like best about the product?
Easy of use and metadata filtering. Pinecone is one of the few products out there that is performant with a query that contains metadata filtering.
What do you dislike about the product?
The pricing doesn't scale well for companies with millions of vectors, especially for p indexes. We experimented with pgvector to move our vectors in a postgres but the metadata filtering performance was not acceptable with the current indexes it supports.
What problems is the product solving and how is that benefiting you?
Semantic search for now.
