DGX Spark Evaluation / Sponsorship Request – Local AI Coding Agents

Accelerated Computing DGX Spark / GB10 User Forum DGX Spark / GB10
, ,
1

Hi NVIDIA Community,

I’m a Linux System Administrator from Denmark, working on a local AI coding infrastructure using NVIDIA GPUs.

My current setup consists of:

  • 2× RTX 3090 (48 GB combined VRAM)
  • Ubuntu Linux, vLLM and llama.cpp
  • Hermes for autonomous coding agents
  • Multiple concurrent AI development sessions

I’m particularly interested in evaluating DGX Spark’s 128 GB unified memory for larger models, long-context inference and parallel coding agents.

I would like to benchmark DGX Spark against my existing dual RTX 3090 setup, including performance, memory utilization, concurrency and power efficiency.

Would anyone from NVIDIA Developer Relations be interested in supporting this project with a DGX Spark evaluation loan or sponsored unit?

In return, I’d be happy to publish reproducible benchmarks, configuration guides and technical findings for the community.

I’d appreciate any guidance on whom to contact.

Thanks!

2

That is highly unlikely to happen.

You can rent DGX Sparks if needed for your research: Scan Cloud NVIDIA DGX Spark Personal AI Supercomputer - 1 Month Access LN168312 - SCAN-SPARK-CLOUD-1M | SCAN UK

3

Agreed. Forum is already full of those benchmarks, config guides, technical findings to the brim.

4

So I have built something I guess that is similar to the original request. I have my apps that I use in a development life cycle, where they are in a loop where they are constantly being fixed, improved or new features when I create requests. I update the stats for this every 6 hours.

Partly an experiment to see what is possible with a local setup, and actual real use to improve the applications that I built for my security tooling.