Built this yesterday with Qwen3.8-Flash-Next (NVFP4, 262K context) on a single NVIDIA DGX Spark

Planning, coding, testing = 8h total. Stack: VSCode Copilot in autopilot mode + SGLang Stats: ∼10k lines generated, ∼800k tokens consumed Sure, it's not GPT-6 Astra level, but for a 100% local ∼180B MoE running on a single DGX Spark at ∼35 tok/s. Not bad...