Build an AI Agent for Enterprise Research Blueprint by NVIDIA | NVIDIA NIM
Looking at the AI Agent blueprint, it seems like the majority of this runs within an H100/80GB setup except for the final LLM for generating the report. I plan to use the blueprint on a DGX Spark with a quantized or smaller final LLM.
Will NVIDIA be updating any of the existing blueprints to target the performance of a locally run DGX Spark GB10 class products, or will it be up to individual end-users to make the modifications as necessary? Thanks.