NVIDIA DGX SPARK

Ollama on a spark is a cheat code to productivity.

While personally I am a huge @grok user on the web, I love to keep the Spark on the side for all kinds of quick references.

ssh into it, start your favorite model (no interface) and just ask one-off questions.

In this case, I https://t.co/xUqGd2jb22

Great news! The latest version of @ollama enables direct integration with Claude Code.

Pull a model you like, run "claude --model <model_name>" and you're all set.

No more faking it with mid layers. 👏 https://t.co/TTl13w1A7k

Small steps for mankind, but big steps for me.

I got Claude Code to run on Nvidia Spark using Qwen3 Next.

Will post detailed video and what it can do tomorrow. https://t.co/pprqPR9FM6

40 TOPS for hardware devices, is it too much or too little? Does anyone have an educated opinion about this?

https://t.co/5cbVKBTAoU