salient-data/qwen38-next-dgx-recipe
Reproducible vLLM recipe: nvidia/Qwen3.8-Flash-Next-NVFP4 on one NVIDIA DGX Spark (GB10) — pinned nightly, patches with provenance, build/launch scripts, results with method
GitHub repository with 10 stars and 1 forks.
Language: Shell
Topics: dgx-spark, dgxspark, gb10, nvfp4, nvidia-dgx-spark, qwen, speculative-decoding, vllm