Build your enterprise Generative AI applications with Microsoft Olive
summary
Microsoft Olive is an open‑source model‑ops tool (now in Windows AI Studio) that runs fine‑tuning with LoRA/QLoRA, converts models to ONNX/GGUF/INT4, and submits Azure ML jobs end‑to‑end for evaluation via Prompt Flow. Demo shows Phi‑series models (Phi‑1 1.3B, Phi‑2 2.7B, Phi‑3 mini variants) served locally with Semantic Kernel, ONNX Runtime and LlamaEdge/wasmedge for GGUF, then deployed to an NVIDIA Jetson Xavier NX (Jetpack 5.01) with low memory and usable latency — all working end to end under Microsoft Olive.
AI-written summary · about
chapters (7)
- 0:00Enterprise GenAI with Microsoft Olive
- 1:18What is SLM and why it matters
- 6:03SLM advantages over LLMs
- 15:06Key SLM application considerations
- 23:11Microsoft Olive for SLM operations
- 30:55Demo: Microsoft Olive in action
- 37:08Summary and SLM future
related talks
45mDeliver AI-powered experiences across cloud and edge with Windows
46mAI made easier: How the ONNX Runtime and Olive toolchain will help you Q&A
23mFoundry Local: Cutting-Edge AI experiences on device with ONNX Runtime/Olive
21mDemocratizing AI with Open Source Multi-Agent Systems
29mOpen Source Sovereign AI: Building Enterprise GenAI Platforms for Control & Comp...
52mSimplifying the adoption of generative AI for enterprises
11mNVIDIA and the AI-Powered Enterprise
38mIntroducing Open Platform for Enterprise AI