COLLECTION / RECIPES

Recipes

One artifact × hardware × engine unit. Validated rows are launch-safe; measured speed sweeps attach to each recipe as evidence.

2,330 records

NVIDIA DGX Spark GB101 × 128 GBNVIDIA-Nemotron-3.5NVFP4sglangdockerUnknown contextcontext—no evidencecandidate sglang docker unmeasured

NVIDIA DGX Spark GB101 × 128 GBNVIDIA-Nemotron-Labs-3-Puzzle-75B-A9B-BF16NVFP4vllmreferenceUnknown contextcontext231.6 tok/smeasuredcandidate vllm evidence measured

NVIDIA DGX Spark GB101 × 128 GBOrnith-1.0-35BGGUF-NVFP4 (qwen35moe)llama.cppreferenceUnknown contextcontext1.5 tok/smeasuredcandidate llama.cpp evidence measured

NVIDIA DGX Spark GB101 × 128 GBOrnith-1.0-35B-AEON-Ultimate-Uncensored-BF16NVFP4vllmreferenceUnknown contextcontext29.1 tok/smeasuredcandidate vllm evidence measured

NVIDIA DGX Spark GB102 × 128 GBOrnith-1.5-35B-A3Bnvfp4vllmdockerUnknown contextcontext—no evidencecandidate vllm docker unmeasured

NVIDIA DGX Spark GB101 × 128 GBQwen3-Next-80B-A3B-InstructNVFP4vllmreferenceUnknown contextcontext39 tok/smeasuredcandidate vllm evidence measured

NVIDIA DGX Spark GB101 × 128 GBQwen3.5-27BUD-Q4_K_XLllama.cppreference256Kcontext11.1 tok/smeasuredcandidate llama.cpp evidence measured

NVIDIA DGX Spark GB101 × 128 GBQwen3.5-27BNVFP4vllmreference4Kcontext11.5 tok/smeasuredcandidate vllm evidence measured

NVIDIA DGX Spark GB101 × 128 GBQwen3.5-35B-A3BUD-Q4_K_XLllama.cppreference256Kcontext60.3 tok/smeasuredcandidate llama.cpp evidence measured

NVIDIA DGX Spark GB101 × 128 GBQwen3.5-35B-A3BUD-Q8_K_XLllama.cppreference256Kcontext35.1 tok/smeasuredcandidate llama.cpp evidence measured

NVIDIA DGX Spark GB101 × 128 GBQwen3.5-35B-A3Bfp8vllmreferenceUnknown contextcontext51.4 tok/smeasuredcandidate vllm evidence measured

NVIDIA DGX Spark GB101 × 128 GBQwen3.5-35B-A3B-BaseGGUF-NVFP4 (qwen35moe)llama.cppreference1Kcontext245.4 tok/smeasuredcandidate llama.cpp evidence measured

NVIDIA DGX Spark GB101 × 128 GBQwen3.5-122B-A10BUD-Q6_K_XLllama.cppreferenceUnknown contextcontext17.1 tok/smeasuredcandidate llama.cpp evidence measured

NVIDIA DGX Spark GB101 × 128 GBQwen3.5-122B-A10BINT4vllmreference128Kcontext33.9 tok/smeasuredcandidate vllm evidence measured

NVIDIA DGX Spark GB101 × 128 GBQwen3.6-27BQ4_K_Mllama.cppreferenceUnknown contextcontext55.2 tok/smeasuredcandidate llama.cpp evidence measured

NVIDIA DGX Spark GB101 × 128 GBQwen3.6-27Bfp8sglangreference64Kcontext4.7 tok/smeasuredcandidate sglang evidence measured

NVIDIA DGX Spark GB101 × 128 GBQwen3.6-27Bfp8vllmreference64Kcontext14.3 tok/smeasuredcandidate vllm evidence measured

NVIDIA DGX Spark GB101 × 128 GBQwen3.6-27BNVFP4vllmreferenceUnknown contextcontext40.1 tok/smeasuredcandidate vllm evidence measured

NVIDIA DGX Spark GB101 × 128 GBQwen3.6-27BNVFP4vllmreference256Kcontext23.5 tok/smeasuredcandidate vllm evidence measured

NVIDIA DGX Spark GB101 × 128 GBQwen3.6-27BNVFP4vllmreference256Kcontext18.9 tok/smeasuredcandidate vllm evidence measured

NVIDIA DGX Spark GB101 × 128 GBQwen3.6-27BNVFP4vllmreference256Kcontext34.9 tok/smeasuredcandidate vllm evidence measured

NVIDIA DGX Spark GB101 × 128 GBQwen3.6-27BNVFP4vllmreference256Kcontext32.8 tok/smeasuredcandidate vllm evidence measured

NVIDIA DGX Spark GB101 × 128 GBQwen3.6-27Bnvfp4vllmdockerUnknown contextcontext—no evidencecandidate vllm docker unmeasured

NVIDIA DGX Spark GB101 × 128 GBQwen3.6-27B-AEON-Ultimate-UncensoredNVFP4vllmreferenceUnknown contextcontext90 tok/smeasuredcandidate vllm evidence measured