Seamlessly automating the journey from training to production, ZenML's new NLP project template offers a comprehensive MLOps solution for teams deploying Huggingface models to AWS Sagemaker endpoints. With its focus on reproducibility, scalability, and best practices, the template simplifies the integration of NLP models into workflows, complete with lineage tracking and various deployment options.
An easy way to deploy an ephemeral MLOps stack, inclusive of ZenML, Kubeflow, MLflow, and Minio Bucket. This one-stop sandbox provides users an interactive playground to explore pre-built pipelines and effortlessly experiment with various MLOps tools, without the burden of infrastructure setup and management.
Explore how ZenML, an MLOps framework, can be used with large language models (LLMs) like GPT-4 to analyze and version data from databases like Supabase. In this case study, we examine the you-tldr.com website, showcasing ZenML pipelines asynchronously processing video data and generating summaries with GPT-4. Understand how to tackle large language model limitations by versioning data and comparing summaries to unlock your data's potential. Learn how this approach can be easily adapted to work with other databases and LLMs, providing flexibility and versatility for your specific needs.
ZenML is launching the ZenML Hub, a novel plugin system that allows users to contribute and consume stack component flavors, pipelines, steps, materializers, and other pieces of code seamlessly in their ML pipelines.
We decided to explore how the emerging technologies around Large Language Models (LLMs) could seamlessly fit into ZenML's MLOps workflows and standards. We created and deployed a Slack bot to provide community support.