howtospark
Recipes

BF16vLLMthinking off

Stub recipe for the BF16 build of Laguna-XS 2.1 (poolside) on DGX Spark via vLLM. Benchmark runs link here; serving budget, steps, and flags are still to be filled in.

Eval scores

(compare all)
91.5HumanEval

Across this model's recipes

HumanEval · median score by recipe

Bench card

No measured runs yet — be the first: install the spark-benchmark skill.

Same quantized model

Overview

_Stub recipe_ — auto-generated so its benchmark runs have a home to attach to. TODO: per-node memory budget, serving steps, key vLLM/llama.cpp flags, and troubleshooting.