Zilliz and AWS are working to provide developers the tools to build modern AI-powered applications. With Zilliz Cloud on AWS, users can quickly scale their vector search storage capacity without the need to re-provision hardware. Running Zilliz Cloud on AWS simplifies integrating your application with other AWS AI and ML technologies.

 <a href="https://aws.amazon.com/marketplace/seller-profile?id=4922a541-e428-480d-8e32-db4ee9a7f46e" target="_blank" class="outlinedButton">Zilliz Offers on AWS</a>



#### Zilliz Cloud on AWS

By subscribing to Zilliz Cloud on the AWS Marketplace, users can take advantage of simplified procurement and consolidated billing to allow developers to spend more time building applications and less time managing vendors and contacts. Zilliz Cloud is available in the US West (Oregon), US East (N. Virginia), US East (Ohio), Europe Central (Frankfurt), and Asia Pacific (Singapore) AWS regions. Visit the AWS Marketplace today to subscribe.

<a href="https://aws.amazon.com/marketplace/pp/prodview-iqbidum7feuio?trk=6291efb7-db60-4f73-9971-842495af9803&sc_channel=el&source=zilliz" target="_blank" class="outlinedButton">Subscribe on AWS Marketplace</a>

Don’t have an AWS account? Try Zilliz for free.

 <a href="https://cloud.zilliz.com/signup" target="_blank" class="outlinedButton">Try Zilliz on AWS for Free</a>

Zilliz partnership with AWS - How you can access Zilliz Cloud on AWS

Together Arize AI and Zilliz help users better understand and fine tune their LLM, CV, and NLP models to increase confidence in their embeddings and the retrieval augmented generation (RAG) and similarity search systems that they build with vector databases like Zilliz Cloud and Milvus. 

#### RAG Evals: Statistical Analysis of Retrieval Strategies

In this video, Jason Lopatecki CEO and Co-Founder and Sally-Ann DeLucia ML Solutions Engineer at Arize AI delve into the 5 pillars of LLM Observability: evaluation, traces & spans, prompt engineering, search & retrieval, and fine-tuning. These pillars explore LLM output evaluations, context retrieval enhancement, and insights on benchmarking and analyzing retrieval systems for LLMs and RAGs.

<a href="https://zilliz.com/event/arize-RAG-evals-statstical-analysis-retrieval-strategies/success" target="_blank" class="outlinedButton">Watch Now</a>

#### Building And Troubleshooting An Advanced LLM Query Engine

​In this on-demand session, go step-by-step on creating a robust query engine using the combined power of Arize Phoenix, LlamaIndex, LangChain, GPT 3.5 Turbo, NLTK, and Milvus. You’ll get an overview of LLM orchestration, an introduction to vector databases and an explanation how search and retrieval works and why it is needed. 


 <a href="https://www.youtube.com/watch?v=_zDDErOaUqc" target="_blank" class="outlinedButton">View Replay</a>

#### Embeddings: Discover the Key To Building AI Applications That Scale with Zilliz
This conference talk focused on using embeddings for scalable generative AI applications. It discusses how the CVP framework can be used to fix a lot of the existing issues around hallucination and the lack of domain knowledge that we see in generative AI models. They see a demo of OSS chat, a manifestation of the CVP framework. 


 <a href="https://arize.com/resource/embeddings-milvus/" target="_blank" class="outlinedButton">Watch Session</a>

#### Extending the Context Window of LLaMA Models Paper Reading
Listen to AI & ML experts discuss a research paper on Position Interpolation (PI), a method extending context window sizes of LLaMA models up to 32,768 positions with minimal fine-tuning. You’ll learn how attention scores work in order to understand what positional embeddings are really there for. 

 <a href="https://www.youtube.com/watch?v=HDm9YjlLE60" target="_blank" class="outlinedButton">View Replay</a>



Zilliz partnership with Arize AI  

Arize AI 

Confluent offers a robust platform for managing data streams in real-time, complete with connectors and native APIs that simplify the integration with other systems. This allows organizations to create custom data pipelines that feed real-time events into vector databases like Zilliz or Milvus, enabling use cases like similarity search, recommendation, and real-time GenAI.

#### Confluent's Kafka Connector for Milvus and Zilliz Cloud: Unlocking the Power of Real-Time AI

The Confluent [Kafka Sink Connector](https://github.com/zilliztech/kafka-connect-milvus#kafka-connect-milvus-connector) with open-source [Milvus](https://milvus.io/) and [Zilliz Cloud](https://zilliz.com/cloud) (the fully managed version of Milvus) enables seamless, real-time vector data streaming from Confluent to Milvus or Zilliz vector databases, unleashing the full potential of unstructured data and significantly enhancing real-time Generative AI powered by large [language models (LLMs)](https://zilliz.com/glossary/large-language-models-(llms)) like OpenAI’s GPT-4.


<a href="https://zilliz.com/blog/announce-confluent-kafka-connector-for-Milvus-and-Zilliz-unlock-power-of-real-time-ai" target="_blank" class="outlinedButton">Read Blog</a>
 
<a href="https://zilliz.com/blog/zilliz-makes-real-time-ai-a-reality-with-confluent" target="_blank" class="outlinedButton">Learn More</a>

#### Build Real-Time RAG Applications Without Hallucination with Zilliz Cloud and Kafka
This integration leverages [Zilliz Cloud](https://zilliz.com/cloud) (Hosted [Milvus](https://zilliz.com/what-is-milvus)) and Confluent Kafka to perform real-time ingestion, parsing, and processing of data to reduce hallucination in LLMs by providing up-to-date and contextually relevant information that helps to enhance the user experience.

<a href="https://zilliz.com/product/integrations/confluent" target="_blank" class="outlinedButton">View Integration</a>




Zilliz partnership with Confluent

Confluent

FiftyOne is the open source machine learning toolset that enables data science teams to improve the performance of their computer vision models by helping them curate high-quality datasets, evaluate models, find mistakes, visualize embeddings, and get to production faster. FiftyOne provides an API to create [Milvus](https://milvus.io/) collections, upload vectors, and run similarity queries, both programmatically in Python and via point-and-click in the app.

#### Search by Natural Language using Milvus Similarity Search

Use Milvus to create a similarity index on your FiftyOne datasets and use this to query your data. The integration documentation provides examples of how to perform some common vector search workflows on a FiftyOne dataset using the Milvus backend such as creating a similarity index. 

<a href="https://docs.voxel51.com/integrations/milvus.html#milvus-setup" target="_blank" class="outlinedButton">View Integration</a>

#### Perform Reverse Image Search using Milvus 

The Reverse Image Search plugin extends FiftyOne’s vector search functionality beyond the data in the dataset. It provides a template for working with local or remote resources, and tailoring FiftyOne to use cases whether you want to build a Google image search clone, or just query popular datasets for dogs similar to your pup.


 <a href="https://voxel51.com/blog/computer-vision-reverse-image-search-plugin-for-fiftyone/" target="_blank" class="outlinedButton">Reverse Image Search Plugin for FiftyOne</a>

#### Unleashing the Potential of Visual Data: Vector Databases in Computer Vision
Explore advanced indexing algorithms and showcase real-world examples in healthcare, retail, finance, and more using the FiftyOne engine combined with the Milvus vector database. 

 <a href="https://voxel51.com/blog/recapping-the-vector-search-themed-computer-vision-meetup-july-13-2023/" target="_blank" class="outlinedButton">Computer Vision Meetup Replay</a>


Zilliz partnership with Voxel51 

Voxel51

Microsoft and Zilliz are working to provide developers with the tools to build modern AI-powered applications. Through integrations with Microsoft Azure, Microsoft OpenAI, and Semantic Kernel, Zilliz and Microsoft are empowering developers with the technologies they need to build next-generation solutions.

#### Zilliz Cloud on Microsoft Azure
Zilliz Cloud is available on [Microsoft Azure](https://azure.microsoft.com/en-us), allowing Azure-centric developers and enterprises to leverage the unique capabilities of Zilliz Cloud, the fully-managed vector database built on Milvus. Zilliz Cloud is currently available in the Azure US East region but has plans to expand to other regions. 

 <a href="https://zilliz.com/blog/zilliz-cloud-now-available-on-microsoft-azure" target="_blank" class="outlinedButton">Read Blog</a>
 
<a href="https://azuremarketplace.microsoft.com/en-us/marketplace/apps/zillizinc1703056661329.zilliz_cloud?tab=Overview" target="_blank" class="outlinedButton">Zilliz on Azure</a>

#### Get Started with Milvus Vector DB in .NET
The .NET team has worked closely with Milvus to enable .NET developers to use vector databases in their applications. Read how to get started using the Milvus .NET SDK with a sample movie recommender app.

 <a href="https://devblogs.microsoft.com/dotnet/get-started-milvus-vector-db-dotnet/" target="_blank" class="outlinedButton">Read Blog</a>

 <a href="https://zilliz.com/product/integrations/.NET" target="_blank" class="outlinedButton">See Integration</a>





Zilliz partnership with Microsoft - How you can access Zilliz Cloud on Microsoft Azure

Microsoft

Airbyte fills the gap left by other data pipeline platforms by focusing on developing and maintaining connectors for a wide range of data sources. Zilliz utilizes Airbyte to enable seamless transfer and processing of data, unlocking new possibilities for real-time, AI-driven applications. Through the integration with [Milvus](https://github.com/milvus-io/milvus) and [Zilliz Cloud](https://zilliz.com/cloud), Airbyte streamlines the process of transferring data and calculating vector embeddings.


#### Use Milvus and Airbyte for Similarity Search on All Your Data

Airbyte is used to facilitate the transfer of data from various sources into [Milvus](https://github.com/milvus-io/milvus), enabling efficient storage, indexing, and similarity searches on high-dimensional vector data. By utilizing embeddings, Milvus/Zilliz can perform similarity searches across unstructured data, leading to improvements in Generative Chat responses, product recommendations, and other applications. This blog shares an example of the creation of a smart support form on a website, where relevant information can be retrieved on the fly to assist users. [Zilliz Cloud](https://zilliz.com/cloud) is leveraged for vector storage, the OpenAI embedding API for calculation, and Streamlit to build a smart submission form showing relevant data.

<a href="https://zilliz.com/blog/use-milvus-and-airbyte-for-similarity-search-on-all-your-data" target="_blank" class="outlinedButton">Read Blog</a>

<a href="https://zilliz.com/product/integrations/Airbyte" target="_blank" class="outlinedButton">See Integration</a>



Zilliz partnership with Airbyte

Airbyte

Zilliz and GCP are working to provide developers with the tools to build modern AI-powered applications. For example, with Zilliz Cloud on GCP, users can quickly scale their vector search storage capacity without re-provision hardware. In addition, running Zilliz Cloud on GCP simplifies integrating your application with other GCP AI and ML technologies.

 <a href="https://cloud.google.com/find-a-partner/partner/zilliz" target="_blank" class="outlinedButton">Google Partner Directory</a>



#### Zilliz Cloud on Google Cloud

Built as a cloud-native vector database solution to power the most innovative search applications, Zilliz Cloud runs on Google Cloud to make it easier to integrate your existing Google technology stack and seamlessly scale your application. By subscribing to Zilliz Cloud on the Google Cloud Marketplace, users can take advantage of simplified procurement and consolidated billing to allow developers to spend more time building applications and less time managing vendors and contacts. Zilliz Cloud is available in the US-West-1 (Oregon) and Asia-Southeast-1 (Singapore) Google Cloud regions. Visit the Google Cloud Marketplace today to subscribe.

<a href="https://console.cloud.google.com/marketplace/product/zilliz-public/zilliz-cloud
" target="_blank" class="outlinedButton">Subscribe on Google Cloud Marketplace</a>

Don’t have a Google Cloud account? Try Zilliz for free.

 <a href="https://cloud.zilliz.com/signup" target="_blank" class="outlinedButton">Try Zilliz on Google Cloud for Free</a>

#### Building for Scale with Zilliz Cloud on GCP

Dopple AI is an innovative platform revolutionizing human-AI interactions. It enables users to create lifelike AI clones, or "Dopples," seamlessly integrating video, audio, and messaging for immersive experiences. Dopple Labs runs Zilliz Cloud on GCP due to their need for large-scale retrieval and the scalability of their tool over time in relation to index size. 

<a href="https://zilliz.com/customers/dopple.ai" target="_blank" class="outlinedButton">Read Case Study</a>



Zilliz partnership with Google Cloud - Access Zilliz Cloud on GCP

Google Cloud

NVIDIA and Zilliz are both active contributors to the Milvus open source vector database community and have worked closely to provide NVIDIA GPU support to Milvus and Zilliz. This collaborative NVIDIA and Zilliz are active contributors to the Milvus open-source vector database community and have worked closely to provide NVIDIA GPU support to Milvus and Zilliz. This collaborative development effort ensures customers realize 10x performance gains over the previous CPU-only version. In addition, NVIDIA and Zilliz deliver robust performance to their users through continued collaboration. development effort ensures that customers can realize 10x performance gains over the previous CPU-only version. Through continued collaboration, NVIDIA and Zilliz are delivering robust performance to their users.

<a href="https://zilliz.com/news/Milvus-2.3-Massively-Accelerates-AI-Powered-Applications-With-NVIDIA-GPU-Support" target="_blank" class="arrowLink">NVIDIA GPU Support Announcement</a>


#### Efficient Vector Similarity Search in Recommender Workflows Using Milvus with NVIDIA Merlin

"NVIDIA Merlin, an open-source framework developed for training end-to-end models to make recommendations at any scale, integrates with an efficient vector database index and search framework. One such framework that has gained much recent attention is Milvus, an open-source vector database created by Zilliz. "

<a href="https://zilliz.com/blog/efficient-vector-similarity-search-recommender-workflows-using-milvus-nvidia-merlin" target="_blank" class="arrowLink">Read More about NVIDIA Merlin and Milvus</a>


#### NVIDIA Announces RAFT Library and GPU Support in Milvus

“Recommender systems use vector databases to store, index, search and retrieve massive datasets of unstructured data. A new important use-case of vector databases is large language models to retrieve domain-specific or proprietary facts that can be queried during text generation. We are introducing a new library, RAFT to accelerate indexing, loading the data, and retrieving a batch of neighbors for a single query. We are bringing the acceleration of RAFT to Meta's open-source FAISS AI Similarity Search, Milvus open source vector database used by over 1,000 organizations and Redis with over 4B docker pulls. Vector databases will be essential for organizations building proprietary large language models.”

**Jensen Huang, CEO, NVIDIA during <a href="https://www.youtube.com/clip/UgkxDb-hOu7zW3WIVyzHbU_kmfD00zvYajlN" class="blackLink">2023 GTC Keynote</a>**


Zilliz partnership with NVIDIA - NVIDIA GPU support to Milvus & Zilliz

NVIDIA

OpenAI provides learning models that convert unstructured data to embeddings and store them in a vector database such as Zilliz Cloud or the open-source Milvus. To help OpenAI users start working with vector data, they partnered with Zilliz to create the OpenAI Cookbook that provides instructions on generating embeddings and storing the data in Milvus.

<a href="https://github.com/openai/openai-cookbook/tree/main/examples/vector_databases/milvus" target="_blank" class="arrowLink">OpenAI Cookbook for Milvus</a>

<a href="https://platform.openai.com/docs/guides/embeddings/limitations-risks" target="_blank" class="arrowLink">OpenAI Documentation for Vector Data</a>



#### Open Source Retrieval Plugins from OpenAI Integrate with Milvus and Zilliz

OpenAI has open sourced [retrieval plugins](https://openai.com/blog/chatgpt-plugins#retrieval) that allow developers to deploy their own plugin and use it with ChatGPT to create a custom knowledge base retrieval solution. The OpenAI embeddings can then be stored in a vector database like Milvus or Zilliz for indexing and searching documents.

 <a href="https://zilliz.com/blog/chatgpt-retrieval-plugin-zilliz-milvus" target="_blank" class="outlinedButton">Learn more</a>

<br/>
<br/>

#### Build a Similarity Search with Zilliz Cloud and OpenAI

This step-by-step tutorial showcases [OpenAI’s Embedding API](https://platform.openai.com/docs/guides/embeddings) with the Zilliz vector database to search across book titles. Many existing book search solutions (such as those used by public libraries, for example) rely on keyword matching rather than a semantic understanding of what the title is actually about. Using a trained model to represent the input data is known as semantic search and can be expanded to a variety of different text-based use cases, including anomaly detection and document search.

 <a href="https://docs.zilliz.com/docs/similarity-search-with-zilliz-cloud-and-openai" target="_blank" class="outlinedButton">Try Tutorial</a>



Zilliz partnership with OpenAI - View OpenAI Cookbook for vector data

OpenAI

Hugging Face is on a mission to democratize good machine learning, one commit at a time. Hugging Face builds open-source resources to empower developers to easily integrate AI into their products and workflows. They believe that AI can be accessible, optimized, and responsible.

<a href="https://huggingface.co/models" target="_blank" class="arrowLink">Explore Model Library</a>



#### Build Question Answering with Hugging Face and Zilliz Cloud

In this step-by-step code tutorial, learn how to build a question-answering system using Zilliz Cloud as the vector database and Hugging Face as the embedding system. 

 <a href="https://docs.zilliz.com/docs/question-answering-using-zilliz-cloud-and-hugging-face" target="_blank" class="outlinedButton">Try Tutorial</a>

Zilliz-Hugging Face partnership - Explore transformer data model repo

Hugging Face

Cohere trains massive language models and offers an API to simplify adding language processing to any system. Through training, this enables users to create huge models customized to their use case and trained on their data. Then, the users generate vectors from this data with a cohere model; it can be ingested into a vector database like Milvus or Zilliz Cloud to build cutting-edge applications and search solutions. 

<a href="https://zilliz.com/product/integrations/cohere" target="_blank" class="arrowLink">Cohere Integrations</a>




#### Build Question Answering Using Cohere and Zilliz Cloud

This hands-on tutorial illustrates creating a question-answering system based on the SQuAD dataset using Zilliz Cloud as the vector database and Cohere as the embedding system.

 <a href="https://docs.zilliz.com/docs/question-answering-using-zilliz-cloud-and-cohere" target="_blank" class="outlinedButton">Try Tutorial</a>

Zilliz partnership with Co:here - Co:here model and integrations

Cohere

LangChain is an AI Agent tool that adds functionality to large language models (LLMs) like GPT. In addition, it includes functionality for token and context management. Together LangChain and Zilliz can be used to inject domain knowledge to build better and more performant LLM applications.

<a href="https://zilliz.com/product/integrations/langchain" class="arrowLink">View Integration</a>

<a href="https://zilliz.com/doc/integrate_with_langchain" target="_blank" class="arrowLink">Getting Started with LangChain</a>

#### Memory for LLM Applications

Connecting external data sources to LLMs to give them memory is crucial to many LLM applications. In this session, LangChain’s [Harrison Chase](https://github.com/hwchase17) dives into retrieval and dissects the challenges of finding relevant information from a large corpus of data.

<a href="https://zilliz.com/event/memory-for-llm-applications-webinar" target="_blank" class="outlinedButton">Watch Replay</a>

Zilliz partnership with LangChain

LangChain

LlamaIndex and Zilliz are partnering to help developers quickly create scalable LLM applications by facilitating intelligent prompting, routing queries, and injecting domain-specific data. LlamaIndex manages LLM interactions by taking data and building an index around it; then, it uses the index to help answer questions related to the data input. Together with a vector database like Milvus or Zilliz, these indexes are stored for faster retrieval and better scalability.

<a href="https://zilliz.com/product/integrations/Llamaindex" target="_blank" class="arrowLink">View Integration</a>

<a href="https://zilliz.com/blog/getting-started-with-llamaindex" target="_blank" class="arrowLink">Getting Started with LlamaIndex</a>

#### Boost your LLM with Private Data using LlamaIndex

In this on-demand session, LlamaIndex co-founder and CEO Jerry Liu shares details on how he built LlamaIndex, the various tools LlamaIndex contains, and how you can get started enriching your LLM with private data. The session ends with an overview of how to use LlamaIndex with Milvus or Zilliz Cloud.

<a href="https://zilliz.com/event/boost-your-llm-with-private-data-using-llamaindex" target="_blank" class="outlinedButton">Watch Session Replay</a>

Zilliz partnership with LlamaIndex

LlamaIndex

PyTorch Foundation is supported by leading contributors to the PyTorch open source project support the PyTorch Foundation. The Foundation leverages the resources members and contributors provide to enable community discussions and collaboration.

<a href="https://pytorch.org/vision/stable/index.html" target="_blank" class="arrowLink">Torchvision Documentation</a>



#### Build Image Search Solution with PyTorch and Zilliz Cloud

Create a simple image search example using PyTorch and Zilliz Cloud. The dataset we are searching through is the Impressionist-Classifier Dataset found on Kaggle. This example uses a 1 CU instance and the Torchvision pre-trained Resnet50 model for embeddings.

 <a href="https://docs.zilliz.com/docs/image-search-with-zilliz-cloud-and-pytorch" target="_blank" class="outlinedButton">Try Tutorial</a>


Zilliz partnership with PyTorch - View image search solution tutorial

PyTorch

Zilliz has partnered with [Voyage AI](https://www.voyageai.com/) to simplify the conversion of unstructured data into searchable vector embeddings in [Zilliz Cloud](https://zilliz.com/cloud) and to help assess the effectiveness of a RAG system implemented with various embedding models for code-related tasks. 

#### Zilliz and Voyage AI: Revolutionizing RAG applications

[Zilliz Cloud Pipelines](https://zilliz.com/zilliz-cloud-pipelines), a tool for converting unstructured data into high-quality embeddings, supports ```voyage-2``` and ```voyage-code-2``` from VoyageAI as embedding models to achieve the highest retrieval quality on code-related tasks. This integration enables scalable API service for retrieval, ideal for use with tools like [LlamaIndex](https://zilliz.com/product/integrations/Llamaindex). Get started crafting superior RAG for code-intensive texts with Zilliz Cloud Pipelines and Voyage AI using the integration below. There is no need to create a separate account for the embedding model, everything is turnkey in your Zilliz Cloud account. 

<a href="https://zilliz.com/product/integrations/voyage-ai" target="_blank" class="outlinedButton">View Integration</a>

#### Crafting Superior RAG for Code-Intensive Texts with Zilliz Cloud Pipelines and Voyage AI

Learn how to assess the effectiveness of a retrieval augmented generation (RAG) system implemented with various embedding models. See how the ```voyage-2``` and ```voyage-code-2``` embedding models perform compared to BAAI and Open AI in terms of retrieval capability on code datasets. This blog digs into the methodology and shows how the Voyage AI models perform significantly better on code-related retrieval tasks.

 <a href="https://zilliz.com/blog/craft-superior-rag-for-code-intensive-texts-with-zcp-and-voyage" target="_blank" class="outlinedButton">Read Blog</a>

Partners

All Partners