After deploying Qwen 3.8-27B locally, I learned another knowledge
Dgx spark is not suitable for running Dense models Different model architectures have different throughput requirements for underlying computing power and memory bandwidth. If a dense model does not do quantification and KV management well, no matter how powerful the hardware is, it will not work. Sure enough, practice brings true knowledge, let’s feel the speed.
QWEN 3.8:27b made this as a one shot prompt for me in 20m!
All while running locally on my machine. Kinda crazy
Agentic Task: Install K3s in your own slicervm
Left: DeepSeek V4 Flash 0731 - 2x DGX Sparks Right: Qwen 3.8 27B (FP8) - 1x RTX 6000 Pro 40 (ish) tok/s gen vs 88 tok/s Why do they both feel "slow"? Thinking aka "variant" - they're spending a lot of tokens on reasoning.
Same prompt: Qwen 3.8 FP8 on the left, Qwen 3.6 FP8 on the right.
The difference in design taste isn't even close. Though it's hard to tell if 3.8 was specifically fine-tuned for this task—the level of polish is almost unreal.
You have reached the end of the archive
All of qwen38