dgx-spark-playbooks

The one thing that's new/worth pointing out are the https://developer.amd.com/playbooks/ (https://github.com/amd/playbooks) - this is AMD's answer to Nvidia's playbooks (https://build.nvidia.com/spark / https://github.com/NVIDIA/dgx-spark-playbooks ) - I think it's great that they're actually taking this more seriously. Hardware is the exact same as what used to be available for $2K last year (and is still $1K cheaper from Chinese OEMs). LTT Lab's LLM testing is getting more sophisticated, which is great - I think it's worth noting that ROCm/Vulkan versions and llama.cpp build versions are going to have some big differences for numbers. For those wanting to get the most out of their Strix Halos, there's both kernel tweaks and utilities like ryzenadj that can help you get the most out of it. ( http://strixhalo.wiki/ has most of that docume

> building something on your desktop that’ll run on data center hardware in production, the DGX Spark is your answer It isn't, because it's a different architecture than the datacenter hardware. They're both called "Blackwell", but that's a lie[1] and you still need "real" datacenter Blackwell card for development work. (For example, you can't configure/tune vLLM on Spark, and then move it into a B200 and even expect it to work, etc.) [1] -- https://github.com/NVIDIA/dgx-spark-playbooks/issues/22

> This is a professional tool for AI R&D. Not really, not it isn't, because it's deliberately gimped and doesn't support the same feature-set as the datacenter GPUs[1]. So as a professional development box to e.g. write CUDA kernels before you burn valuable B200 time it's completely useless. You're much better off getting an RTX 6000 or two, which is also gimped, but at least is much faster. [1] -- https://github.com/NVIDIA/dgx-spark-playbooks/issues/22

Update: The very first Playbook I tried (unsloth) didn’t work . There was a bad URL and some package version and dependency issues. I figured out how to get it working and completed the playbook, then submitted a PR that fixes the playbook instructions for others. The PR has not been merge as I write this, so if you run into a problem with the Unsloth playbook, see: PR #14: https://github.com/NVIDIA/dgx-spark-playbooks/pull/14