20:59
2026-09-10
github.com
large-language-models
DeepSeek-v4-flash-0731-spark-sparkinfer: DeepSeek V4 Flash on one DGX Spark
A pinned Docker recipe now serves the 0xSero/deepseek-v4-flash-0731-spark model on a single NVIDIA DGX Spark via Local Inference Lab's SparkInfer, exposing a 262,144-token limit with a K64 DSpark spec…