We unlocked NVFP4 on the DGX Spark: 20% faster than AWQ!

[griffith.mark]

Nice Job!! Looking forward to implementing the model myself. I do really like the Qwen3 Next Instruct 80b model short TTFT vs Nemotron Nano etc which when I am working fit better with my work flow than than waiting 30-45s in thinking mode.

To get to 60-120 toks/ sec from my baseline of 42 toks/sec using regular images is very impressive.

DGX Spark + Qwen3-Next-80B: Proven Performance, But Missing Clear Path to NIM, TensorRT-LLM & Web UIs

DGX Spark + Qwen3-Next-80B: Proven Performance, But Missing Clear Path to NIM, TensorRT-LLM & Web UIs Hi NVIDIA team and fellow DGX Spark owners, After about six weeks with DGX Spark, coming from a macOS / non-Linux background, this box genuinely feels like an “iPhone moment” for local AI: a compact, headless powerhouse that runs 80B-class models like Qwen3-Next-80B-A3B-Thinking at usable speeds on a desk, no datacenter required. The overall experience has been overwhelmingly positive — but m…

This will change the game for NVFP4 on DGXspark.

Many thanks,

Mark