hsm2078546 byronvoorbach5512 I m

aarminta7

@msh702 @Byron Voorbach {Weaviate} I'm thinking of switching to Weaviate. What would be some pros and cons for my use case over something like Pinecone? Feel free to DM.

9 comments

mmsh702

hey @arminta7 what is your use case?

aarminta7

Querying personal research and notes.

mmsh702

Hey @arminta7 ,

I can’t speak for Pinecone, but these are some of the main points where we feel like weaviate distinguishes itself from other vector databases:

mmsh702

Open source

Weaviate is fully open source. Whether your run Weaviate yourself or on our Weaviate Cloud Service (our managed service of weaviate), there is no limitation in features. In addition to benefiting from code contributions and support from the community, this makes starting with vector search very accessible to many companies.

mmsh702

Module system

Weaviate is completely modularised. Without any modules attached, weaviate is a pure vector-native database and search engine. Modules allow users to extend weaviate to other use cases. For example, the txt2vec-transformers(https://weaviate.io/developers/weaviate/current/retriever-vectorizer-modules/text2vec-transformers.html), qna-transformers (https://weaviate.io/developers/weaviate/current/reader-generator-modules/qna-transformers.html) , and sum-transformers (https://weaviate.io/developers/weaviate/current/reader-generator-modules/sum-transformers.html) grants weaviate the ability to do text vectorization, Q&A, and text summarisation, respectively. With more modules on the horizon!

mmsh702

Partner integrations

Weaviate has out-of-the-box integration with many popular machine-learning frameworks/services. For example, you can use Hugging Face’s Inference API(https://weaviate.io/developers/weaviate/current/retriever-vectorizer-modules/text2vec-huggingface.html) to vectorize your data or Open AI’s Completions endpoint(https://weaviate.io/developers/weaviate/current/reader-generator-modules/qna-openai.html) to do Q&A. Weaviate is also connected through multiple vector search frameworks, like haystack(https://github.com/deepset-ai/haystack), jina(https://github.com/jina-ai/jina) and txtai(https://github.com/neuml/txtai).

mmsh702

UX focus

We strive to make it as pleasant and easy as possible for users to start working with Weaviate. This ranges from documentation to 3rd party modules to operationalizing the database. For example, you can quickly create your own customized weaviate instance using our docker compose configurator(https://weaviate.io/developers/weaviate/current/installation/docker-compose.html) or our helm chart(https://weaviate.io/developers/weaviate/current/installation/kubernetes.html) if you want to deploy it to Kubernetes.

mmsh702

Community driven roadmap

We really value community at SeMI and work very hard to integrate that into our development roadmap. We encourage our community to participate in our roadmap by raising and upvoting feature requests. You can have a look at our feature roadmap here: https://weaviate.io/developers/weaviate/current/roadmap/feature-roadmap.html

mmsh702

I hope this answers your question! Please let me know if you’d like to hear more about Weaviate.

Add a reply

Find answers from the community

hsm2078546 byronvoorbach5512 I m