Skip to main content

Product Forums

This dedicated space for users of HP Z product users to seek assistance, share insights, and collaborate with fellow members.

25 Topics
Unlock Early Access to Z by HP Boost – Sign Up Now
Discover how Z by HP Boost puts the power of composable computing in the hands of AI and ML professionals, offering GPU resources that adjust seamlessly to your workflow. Z by HP Boost allows you to dynamically manage GPU power, scaling up or down based on your needs. Whether you're training intricate models, running demanding simulations, or working on data-heavy projects, this flexible architecture ensures you get the right compute power, exactly when you need it. Why should your compute stay the same when every workflow is different? With Z by HP Boost, we're not just enhancing productivity—we’re reshaping how AI and ML workflows are optimized. Interested in transforming your workflow? Join our Early Access Program and be part of the journey. Your insights will help shape the future of this product. Sign up for the waitlist today! Register here for the Z by HP Boost Early Access Program By providing on-demand access to high-performance workstation GPUs, it allows firms to efficiently maximise their resources, utilising hardware that might otherwise sit idle under a desk, especially at night. The alternative is equipping everyone with high-end GPUs or running everything in the cloud. Both options are expensive and cloud can also bring unpredictable costs. Keeping things local also helps firms protect intellectual property, keeping proprietary designs and the models that are trained on their proprietary designs behind the firewall. - AECMag
Contained AI, Protected Enterprise: How Containerization Allows Developers to Safely Work with DeepSeek Locally
Introduction Organizations face a critical challenge when using AI: balancing the need for powerful AI capabilities with stringent data privacy requirements. As AI models become more sophisticated, the complexity of managing development environments while ensuring data security has become a significant hurdle for many teams. AI engineers and data scientists want to work with the latest cutting-edge models, whether they be open source or proprietary LLMs, but often find themselves spending valuable time configuring environments and wrestling with dependency conflicts, all while trying to maintain the security standards their organizations demand. Enter HP AI Studio and it’s 1-click containerization, a solution that transforms how developers work with sophisticated AI models like the recently released DeepSeek models in a secure, local environment. This innovation directly addresses the dual challenges of development environment management and data privacy, allowing teams to focus on what matters most – building exceptional AI applications. In this article, you'll learn how to leverage HP AI Studio's containerization capabilities to create reproducible, secure development environments for your AI projects in seconds, not hours. We'll walk through the entire process, from initial setup to advanced implementation, showing you how to save valuable development time while maintaining the highest standards of data privacy. Technical Background The Environment Challenge The modern AI development stack is inherently complex, requiring careful orchestration of multiple dependencies, libraries, and model weights. Traditional approaches to environment management, such as virtual environments, often fall short when dealing with the specific requirements of large language models like the newly released DeepSeek series of models. Developers frequently encounter version conflicts, incompatible dependencies, and inconsistent behavior across different machines. Current Limitations Traditional approaches to managing AI development environments typically involve: Manual virtual environment creation and management Complex dependency resolution processes Inconsistent environment reproduction across team members Time-consuming setup procedures for each new project Limited isolation between the host system and AI development environment These limitations not only slow down development but also create security vulnerabilities when working with sensitive data and powerful AI models, potentially exposing local host machine file systems and corporate network processes. Setting Up Your Containerized Environment in HP AI Studio HP AI Studio simplifies AI developer workflows by moving beyond virtual environments and creating containerized development environments. Normally this would involve an equally cumbersome Docker setup, but with HP AI Studio we’ve condensed this down into a single click. Downloading DeepSeek to my local machine To begin using DeepSeek within an HP AI Studio containerized environment go to HuggingFace and select the specific model repo you’d like to use. For the purposes of this demonstration I used the DeepSeek-R1-Distill-Qwen-1.5B (deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B · Hugging Face). Save the model files somewhere on your local machine where you can access them later. I’ve also included code (See implementation below) that will allow you to download during the implementation phase of the project. Basic Setup Process in HP AI Studio Selecting a pre-configured workspace With HP AI Studio we provide a number of pre-configured development images that contain all the tools you need for your different AI workflows. Once you’ve selected one of these, you are ready to build your containerized development environment. I’ve selected the ‘Local GenAI’ image because it provides me common GenAI libraries, e.g., Langchain, llama.cpp, and local vector database packages. Getting DeepSeek model files into HP AI Studio From here, I can simply point to the location on my local machine where I downloaded and saved the DeepSeek model and configuration files from HuggingFace. You don’t point to specific files, just the local directory itself in the ‘Local Folder’ section. Container Configuration Next comes the all-important containerization step which will isolate my local development environment on my machine. By simply pressing the ‘Play’ button highlighted above AI Studio builds the container for you. The container automatically includes: All necessary Python dependencies Optimized environment variables Security configurations to ensure that processes and model execution taking place inside the container does not impact anything on a local host machine Practical Example: Building a secure AI chatbot Let's walk through building a privacy-preserving RAG based application using DeepSeek in our containerized environment. As helpful context, I work as a product manager at HP so I wanted to set up a demonstration that resonated with me, but you could replace the dataset I’ve used within any dataset that it useful for you. The dataset I’m using is a from blog post around different growth tactics a product manager might use to help grow the core business for a software product. Here you’ll find the code I used to build the RAG based system, as well as some AI evaluations of my system using Galileo, an HP partner and industry leader in AI evaluation and observability (What is Galileo? - Galileo) Implementation Steps ### Load and install any additional libraries needed import os !pip install PyPDF2 !pip install faiss-gpu !pip install langchain_community[KH1] [CB2] from urllib.request import urlretrieve import numpy as np from langchain_community.embeddings import HuggingFaceBgeEmbeddings from langchain_community.llms import HuggingFacePipeline from langchain_community.document_loaders import PyPDFLoader from langchain_community.document_loaders import PyPDFDirectoryLoader from langchain.text_splitter import RecursiveCharacterTextSplitter from langchain_community.vectorstores import FAISS from langchain.chains import RetrievalQA from langchain.prompts import PromptTemplate import PyPDF2 import faiss from huggingface_hub import snapshot_download ###Download the DeepSeek and embedding models locally to AI Studiolocal_model_dir = "/datafabric/DeepSeek-Distill-Qwen1.5B” local_embed_dir = "/home/jovyan/datafabric/mxai-embed", repo_id = 'deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B' #DeepSeek model embed_repo_id = ‘mixedbread-ai/mxbai-embed-large-v1’ #embedding model snapshot_download(repo_id=repo_id, local_dir=local_model_dir,repo_type="model") snapshot_download(repo_id =embed_repo_id, local_dir=local_embed_dir,repo_type=”model”) ### Bring in a PDF data source and set up code to collate pages if PDF extends across multiple pages pdfFileObject = open("growth_loops_product_school.pdf", 'rb') pdfReader = PyPDF2.PdfReader(pdfFileObject) count = pdfReader.pages full_doc = '' for i in range(len(count)): page = pdfReader.pages[i] full_doc += page.extract_text() ### Previous output 'full doc' is just one big string and a 'Document' object ### is required for some later langchain calls. So, define a function that creates ### the Document object def generate_tokens(s): text_splitter = RecursiveCharacterTextSplitter(chunk_size=500, chunk_overlap=50) splits = text_splitter.split_text(s) return text_splitter.create_documents(splits) ### Create document chunks docs_after_split = generate_tokens(full_doc) ### Encode the document chunks using an embedding model I also brought into HP AI Studio huggingface_embeddings = HuggingFaceBgeEmbeddings( model_name="/home/jovyan/datafabric/mxai-embed", model_kwargs={'device':'gpu'}, encode_kwargs={'normalize_embeddings': True}, ) ### Set up local vector store and retriever vectorstore = FAISS.from_documents(docs_after_split, huggingface_embeddings) retriever = vectorstore.as_retriever(search_type="similarity", search_kwargs={"k": 4}) ### Set up DeepSeek model pipeline from transformers import pipeline pipe = pipeline(task = "text-generation", model='/home/jovyan/datafabric/DeepSeek-Distill-Qwen1.5B', temperature = 0.15, max_new_tokens = 300, device = -1) from langchain_community.llms.huggingface_pipeline import HuggingFacePipeline hf = HuggingFacePipeline(pipeline=pipe) llm = hf ### Set up a prompt template which will be used with the LLM from langchain.prompts import ChatPromptTemplate from langchain.schema import StrOutputParser from langchain.schema.runnable import RunnablePassthrough from typing import List from langchain.schema.document import Document def format_docs(docs: List[Document]) -> str: return "\n\n".join([d.page_content for d in docs]) prompt_template = """ {context} Question: {question} Helpful Answer: """ PROMPT = ChatPromptTemplate.from_template(prompt_template) chain = {"context": retriever | format_docs, "question": RunnablePassthrough()} | PROMPT | llm | StrOutputParser() ### Set up model evaluation using Galileo (HP Partner for AI Evals) import promptquality as pq os.environ['GALILEO_API_KEY'] = "your api Key" galileo_url = "https://www.galileo.io/" pq.login(galileo_url) Scorers = [pq.Scorers.context_adherence_luna, pq.Scorers.correctness, pq.Scorers.pii, pq.Scorers.completeness_luna, pq.Scorers.chunk_attribution_utilization_luna] prompt_handler = pq.GalileoPromptCallback( project_name="growth_RAG", scorers=Scorers ) ### Run your chain experiments across multiple inputs with the galileo callback inputs = [ "Why are growth loops important?", "What kind of growth loops does Dropbox use?", "What are some limitations of the AARRR framework?", "What are network effects?" chain.batch(inputs, config=dict(callbacks=[prompt_handler])) ### Publish the results of your run to Galileo prompt_handler.finish()``` Expected Outputs I won’t go over all the expected outputs but using the last question I provide the LLM ‘What are network effects?’, I would expect a basic explanation of what network effects are in the context of growing a software product and why they are important. Using DeepSeek I was able to get exactly that from the dataset I provided it. While I need to potentially test model performance further, I feel confident that using DeepSeek within HP AI Studio I have built a basic RAG based system that provides answers I can be confident in, all while protecting my local machine from any potential security concerns of using this cutting-edge open source LLM. Conclusion and Additional Considerations HP AI Studio's 1-click containerization revolutionizes how teams work with sophisticated AI models like DeepSeek. By eliminating the complexity of environment setup and management, teams can save significant time while maintaining robust security standards. The ability to create consistent, isolated development environments with a single click not only accelerates development but also ensures that privacy and security remain at the forefront of AI implementation. Constainers vs. Virtual Environment Containers offer several key advantages over virtual environments: Complete isolation: Full system-level isolation ensures better security Reproducibility: Guaranteed consistent environment across all deployments Resource efficiency: Better resource management than virtual machines Version control: Easier to track and manage environment changes Scalability: Seamless scaling from development to production Ready to transform your own AI development workflow? Contact HP's AI Solutions team to learn more about how HP AI Studio can help your organization build secure, efficient AI applications. You can find contact information here Z by HP AI Studio | HP® Official Site to schedule a demonstration.
Introducing Z by HP Boost! 🚀
Announced at #HPImagine , Z by HP Boost enables remote GPU sharing between devices and host workstations, turning idle GPUs into powerful, on-demand resources that deliver performance precisely when it’s needed most. Boost Your Team’s Impact Provide remote access to GPUs in workstations you own and turn PCs into powerhouse devices. Deliver power for teams to iterate quickly, improve model development, and deliver AI solutions effectively. Remote GPU Access. Anytime Productivity. Empower teams to tackle complex AI workflows from nearly anywhere. With remote GPU access, teams can train AI models on their own devices without compromising on compute. Unlock Idle Resources Grant access to single or multiple GPUs within your Z by HP workstations and give individuals access to compute when needed or free it up for others when they don’t. Harness the full potential of your GPUs today and in the future. Secure Data Protection Backed by a leader in security, Z by HP ensures protocols are in place to protect your data. By working with data within your infrastructure, risks associated with cloud processing are minimized. Interested in Learning More About Boost? Visit hp.com/boost for detailed information, and don’t miss the opportunity to join our exclusive webinar on October 4th ! Looking to get involved in our Early Access Program? We invite you to register for our Early Access Program to experience Z by HP Boost firsthand. Your feedback will play a crucial role in shaping the future of this product. Join the waitlist here and be considered for the program.
🎥 Skip the Cloud: Train AI Models Using Local Resources with AI Studio (Webinar Recap)
Thank you for joining our recent webinar, “ Skip the Cloud: Train AI Models Using Local Resources with AI Studio ”, and hope you found the session both engaging and informative! If you’d like to revisit the webinar or explore additional resources, you can find the recording and related materials below. Additionally, if you're interested in learning more about AI Studio, feel free to schedule a one-on-one session with us here: https://calendly.com/sothan-thach-hp/ 🎥Webinar Recording We're looking forward to connecting with you at our next webinar on October 4th (Register Here!) . Also get ready for an exclusive look at the latest innovations from HP and the Z by HP team, coming your way at HP Imagine 2024 on September 24th. Stay tuned for all the exciting updates! Resources 🖥️ Z by HP AI Studio 📝 AI Studio EAP Sign-up Form Webinar Q&As Q: Are there any specific requirements for data preparation before uploading an asset from local storage or S3? A: That's a great question. So, this isn’t about data management or pipeline processing functionality. When you update or attach a pointer to a new asset, you’re essentially just pointing to a folder. Your assets or data are treated as objects. If you need to create a data pipeline, that would have to be done programmatically—either through your notebook or via your usual processes to bring in the data for management. Q: Does HP handle the syncing between all peers in the background? A: Yes, absolutely. Peer-to-peer syncing is handled by a mechanism called SyncThing. This allows, for example, if I open a project and place something in the shared folder in the notebook, it will automatically sync between me and my collaborators. This syncing is peer-to-peer and helps facilitate quick object sharing among team members. You can turn off this sync feature if needed, but it’s useful for fast collaboration. As for metadata syncing, that's managed through MongoDB, hosted by HP. It handles metadata about the account and definitions, including data pointers. For instance, if you or a collaborator point to a specific dataset, but they don't have the right permissions (e.g., access to the S3 bucket), your data policies will prevent them from using it. The endpoint will effectively be locked, and they won’t have access. HP’s OneCloud service manages the metadata syncing in the background. Q: Will you be adding support for additional NVIDIA GPU drivers for Tesla K80 GPUs? A: AI Studio is designed to be hardware- and OS-agnostic. Our vision is to empower data scientists who work across various hardware and operating systems by providing the tools they need. What we aim to facilitate is easy access to the compute resources they require, including GPUs. Currently, AI Studio supports NVIDIA GPUs with drivers version 528.89 or higher. If Tesla K80 is compatible with those drivers, then it's already supported Q: Will HP support Azure AI Studio development environments, and do you plan to offer your own compute resources in the future? A: First, Azure is definitely part of our roadmap. Our goal is to help you access the compute power you need, no matter what tools, operating systems, or hardware you're using. That’s the vision. We already have integrations with several Microsoft products—Windows, WSL 2, GitHub, and Azure Blob Storage are a few examples. Upcoming additions on the roadmap include VS Code, Azure cloud services, AWS cloud services, and more. We've listened to feedback from our users, and they appreciate having access to local GPUs for training and development. However, sometimes local compute isn’t enough, and they need additional power. The idea is to allow users to seamlessly transfer their projects and configurations to another environment with greater compute resources—whether that’s an on-premise server or a cloud service. Ultimately, when you're ready to deploy or train, you should be able to choose where and how that happens—whether it's on your local hardware or in the cloud. The goal is to give you the flexibility to work wherever you need. Q: Do you plan to support community-developed integrations to expand the capabilities of AI Studio? A: Currently, while you can share your applications or technologies with your colleagues through the community, we don’t have a formal marketplace for sharing libraries at this time. However, it’s an interesting idea, and we’re definitely considering it for the future. Q: Does AI Studio have a Vision Language Model (VLM) that can recognize and label elements in architectural blueprints or plans, such as walls, doors, and other features? A: Right now, we're working on adding more integrations to support automation for tasks like labeling, annotation, and data preparation. Our goal is to let you use the tools you’re already familiar with, rather than trying to replace them, while still providing the compute resources you need to get the job done. Currently, we don’t have these additional integrations available. However, we’ve heard your feedback loud and clear and recognize the need for more functionality. We understand that integrations with cloud compute deployments, Microsoft Azure, NVIDIA, and tools for labeling, annotation, and data processing are crucial. At the moment, you can still implement these integrations programmatically through notebooks, but we don’t have a simplified, no-code API or user interface for these tasks yet. Q: Will we have the option to integrate open-source models into our projects during the project setup? A: Yes, we are currently developing that capability. In the next release of AI Studio, we’ll introduce integration with NVIDIA’s catalog, which includes GPU-optimized models. This catalog features both open-source and NVIDIA-vetted models. AI Studio users will have immediate access to these NVIDIA assets, including free models for transfer learning. Additionally, we’re working on integrating with other popular model hubs, which should be available in the next release or soon after. Q: Is the end goal for AI Studio to become API-agnostic, allowing integration with third-party development tools and the Ollama Web UI? A: Yes, you can integrate open-source models into your projects during setup. Currently, in the custom environment, you can create a workspace and use base images provided by us to get started quickly. You can specify whether you're using CPU, GPU, or in the future, NPU. Within this environment, you can add and install libraries, such as LLaMA, using package management tools like Conda, Pip, and Git. While it can be tricky to ensure everything works seamlessly due to the containerization, we provide the flexibility to install and work with libraries using these tools. Q: In the upcoming release, which Vision Language Models (VLMs) are prioritized on your roadmap? Specifically, will models like Microsoft's Florence-2, PaliGemma, YOLO, and SAM 2 be included? A: YOLO is definitely supported in our AI Studio. We’re focusing on three main use cases: traditional machine learning, deep learning, and computer vision. YOLO, being a popular tool for computer vision, is included in the packages and tools we support. Regarding Generative AI improvements and additional frameworks or models, we are integrating tools like LLaMA and working on prompt quality improvements with our partner, Galileo. You can also create a custom workspace in AI Studio and include packages like YOLO as needed. As long as you configure your container properly, you should be able to use these tools effectively.

Cookie policy

We use cookies to enhance and personalize your experience. If you accept you agree to our full cookie policy. Learn more about our cookies.

 
Cookie settings