Popular repositories Loading

1. dgx-spark-qwen38 dgx-spark-qwen38 Public

   Fastest measured Qwen3.8-27B config for DGX Spark (GB10): SGLang + NVFP4 + DFlash2, 72 tok/s greedy median, 1M context, deterministic boots. Flash-Next 176B on one box. One command, everything pinn…

   Python 187 12

2. talkink talkink Public

   Talk. Ink. Push-to-talk dictation for macOS, 100% on-device. Pick your model: Qwen3-ASR, NVIDIA Nemotron or Voxtral, all via Apple MLX.

   Swift 8

3. BTC-Puzzle-Solver BTC-Puzzle-Solver Public

   Python 2

4. qwen3.8-27b-in-16gb qwen3.8-27b-in-16gb Public

   Qwen3.8-27B, 142k context, one 16 GB GPU, one command.

   Shell 2 1

5. btc-puzzle-solver-python btc-puzzle-solver-python Public

   Python 1

6. local_vibe_coding_dgx_spark local_vibe_coding_dgx_spark Public

   DGX Spark Local Inference Toolkit: llama.cpp, Vibe-Coding setup, and Whisper TUI.

   HTML 1