vcruz305/DSV4.1-Flash-1017-EXL3-Spark-recipe
DSV4.1-Flash-1017 SAGE EXL3 (1.62 bpw) on one DGX Spark with vLLM: DSpark MTP, cold-expert ATS alias, tile-K 16 MoE, memory guard, measured 31-34 tok/s
GitHub repository with 32 stars and 4 forks.
Language: Python