8 comments

  • ronef a minute ago

    Huge plus to anyone interested in the space to check out what Graham built here!

    If anyone is also interested on Nix/CUDA/Capital Markets/Flox, we recently did another case study in the space - https://flox.dev/blog/deploying-hardened-flox-nvidia-cuda-st...

  • thenobsta 7 minutes ago

    This is incredible. I have a Jetson lying around and will try to it out on this. I use it to play with vision models, not LLMs, and have been wanting a better way to manage the machine.

  • haunter 5 minutes ago

    Thanks for sharing, saving this for when I get a DGX Spark

  • redrove an hour ago

    Been running this on a few Asus GX10 machines with k3s on top, it’s been great. I’m running the new deepseek.

    Thank you for your work!

    • pixelesque 36 minutes ago

      What quant are you using, and what tps are you getting with K3?

      • redrove 5 minutes ago

        The FP8 version from DeepSeek themselves [0], around 1800 tps prefill and 45 tokens per second decode.

        I’ve been running a custom VLLM image with b12x as well as nvfp4_ds_mla.

        I would say it’s quite fantastic in day to day, I use it mostly in Hermes and sometimes for coding.

        I have qwen 3.6 27b in an rtx 6000 pro as well so I use that as a workhorse in pi with DS as a reviewer/planner.

        [0] https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash-0731

        Edit: I think you may have misread my post. k3s is NOT kimi k3, and I did mention I was running deepseek.

  • nixie-tubes 2 hours ago

    This has been amazingly helpful for managing my DGX Spark! Thank you for all your time and effort into this project!

    • graham33 2 hours ago

      Good to hear, thanks!