SAP Community will be read-only this weekend (Saturday, August 31, through Sunday, September 1). Activities requiring log-in won't be possible. Learn more in this What's New post.
In the blog post about Function Calling LLMs for SAP: Structured Outputs and API calling by @MarioDeFelipe, Mario has done a great job in explaining the why and how of LLM's structured output with Function Calling for SAP API integration.On 6 Aug 202...
Following the blog post Product Updates for SAP Build Code – July 2024 Edition by @BirgitS. Use Joule to enhance your already existing data models and services is now supported in SAP Build Code, which enables the developers to enhance a specific dat...
Authors: @YatseaLi, @cesarecalabria, @amagnani, @jacobahtan In my previous blog post about Bring Open-Source or or Open-Weight LLMs into SAP AI Core, we have gone through an overview introduction of deploying and running open-source LLMs in SAP AI Co...
In my previous blog post about Bring Open-Source or or Open-Weight LLMs into SAP AI Core, we have gone through an overview introduction of deploying and running open-source LLMs in SAP AI Core with BYOM approach, the use cases of open-source LLMs, an...
The open-source community surrounding Large Language Models (LLMs) is evolving rapidly, with new models(model architectures), backends(inference engines), libraries, and tooling constantly emerging. These developments enable the running of LLMs local...
Update on 25 July 2024:Function calls are now supported most recent mistral-v0.3 and llama3.1, which is essential for reliable application integration with llms in native json mode. For more details about function call in LLM, here is a great blog po...
As of 9 July 2024, Concurrency is now enabled by default since Ollama v0.2. This unlocks 2 major features:Parallel requests Run multiple models as long as enough GPU memorySource: https://x.com/ollama/status/1810480544976626159
Great news from SAP Sappire 2024. SAP Business AI Innovations and Collaborations with Partners including Google Cloud, Meta, Microsoft, Mistral AI and NVIDIA Ignite Ingenuity.The leading open-source models LLaMa 3-70b and Mixtral are now available in...
Hi @ido_perez, Thanks for your feedback.I am able to run llama3-70b (Q4_0, size as 40G) with resource plan on infer.l(T4 - 16G GPU VRAM) and train.l(V100 - 16 G GPU VRAM) in SAP AI Core. However, it is very slow, taking 2~3 minutes to respond to a si...