IBM Developer

Introducing IBM Granite.Code A single, lightweight extension that is built to work with IBM's cutting-edge Granite Large Language Model (LLM) for code and is available to individual developers for free! IBM Granite.Code provides robust, context-aware AI support directly in the developer's environment to provide explanation, documentation, translation, and unit test generation Read our latest blog to learn about this powerful AI coding companion: https://ibm.co/3TizKlM

It is crucial yet tedious task to compare outputs from multiple LLMs for selecting a model that is tailored to your specific need. A model that maximize performance, minimize biases, and optimize costs. It can be tricky to choose between open-source models like Llama, Granite, and Mistral. To facilitate this process, Senior GenAI Engineer at IBM, Bhavishya Pandit, has authored this tutorial to provide solutions for a systematic evaluation and the code necessary to compare various LLMs, highlighting the benefits of efficient model selection and optimized workflows: https://ibm.co/4hs2seM #AI #LLM #OpenSource

Large Language Models (LLMs) have immense potential, but they come with challenges like the need for high-quality training data, specialized skills, and extensive computing resources. Forking and retraining these models can be time-consuming and costly. That’s where InstructLab steps in! In this tutorial, CSM Solutions Architect at IBM, Ahmed Azraq explains how to accomplish the same goal of fine-tuning open-source LLMs using InstructLab UI: https://ibm.co/4fmmR2V The InstructLab User Interface (UI) allows you to easily contribute knowledge or skills to the InstructLab taxonomy repository without worrying about YAML structure, the different validation rules, or the GitHub pull request (PR) process.

Explore this tutorial that explains the use of two approaches for migrating a NIM server to the latest AIX release using the virtual media library and the alt_disk_mksysb command. Based on the NIM resources and the infrastructure, you can choose any of the methods for migrating your NIM server. By following these procedures, you can smoothly migrate your NIM server while minimizing risks to your environment: https://ibm.co/4fmnVDP

Learn how to leverage environmental data in your applications through a structured learning path. Click on this link to start your learning journey now: https://ibm.co/48qiLVl

Discover IBM Environmental Intelligence that provides powerful APIs, geospatial data, and environmental insights to build innovative and sustainable applications. Learn how to optimize energy usage, assess carbon footprints, analyze wildfire risk, and predict climate impacts with cutting-edge tools for climate resilience and environmental risk management: https://ibm.co/4e5jnk8

Ever wondered what it’s like to have an AI assistant that writes code, debugs, and handles documentation—all locally on your machine? AI Engineer Kelly Abuelsaad shows you how to use a local open source AI code assistant to develop applications, powered by the Granite Code model. If you're looking to integrate AI into your development process while maintaining privacy, control, and open source compatibility, this guide shows you how to get started! https://lnkd.in/gRCSVvnx #AI #LLM #opensource

Granite models are open, state-of-the-art AI models that provide powerful performance, super safety, and top-notch security all in one family. IBM has announced the release of Granite 3.0, the 3rd-generation of Granite flagship models, designed as 'workhorse' models for generative AI, delivering strong performance for tasks such as RAG, classification, summarization, entity extraction, and tool use. The Granite models are released under the permissive Apache 2.0 license, making them unique in the combination of performance, flexibility, and autonomy they provide to enterprise clients and the community at large. Learn more about how you can get started right away with the IBM Granite models. #Granite #AI #LLM https://lnkd.in/gKBB43sg

If you are going to the IBM TechXchange conference in Las Vegas, be sure to attend these 90 minute hands-on workshops. Bradley Steinfeld and James Reeve will share with you the latest techniques for building AI apps. These are the same techniques we rely on to build #AI for the IBM Skills Network. This is not you typical presentation; you are actually going to build AI so make sure to bring your laptop. Not going to the conference? No problem. We are going to publish these projects on Cognitive Class for you to do these on your own at your convenience. #GenAI

In the evolving world of AI and language models, ensuring that outputs are factually accurate and relevant is crucial. Developer Experience & DevRel Leader, Roy Derks has authored this tutorial that explains the implementation of LLM guardrails in watsonx Flows Engine to ensure the reliability of your RAG applications: https://ibm.co/48aS0Ur By leveraging contextual grounding checks, scoring metrics, and customizable models, you can fine-tune your flows to provide accurate, relevant, and grounded responses. Combined with the flexible, low-code nature of watsonx Flows Engine, this added layer of safety minimizes hallucinations, making AI-driven applications more trustworthy for end users.