A curated list of tools, guides, playbooks, and resources for the NVIDIA DGX Spark (GB10 Grace Blackwell personal AI supercomputer).

vLLM recipe: Xiaomi MiMo-V2.5 + DFlash speculative decoding + fp8 KV cache on 2x DGX Spark (GB10/sm_121)

GLM-5.2-Int4-Int8 on 8x GB10 cluster

GLM-5.2 (744B/40B MoE) on a 4× DGX Spark / GB10 (sm_121) cluster: portable Triton sparse-MLA kernels, a data-free expert prune, MTP draft, and a one-script bootstrap.

Recipe: GLM-5.2 (unpruned QuantTrio Int4-Int8Mix) at 655,360-token context + MTP spec decode (DCP4, fp8_ds_mla) on a 4x NVIDIA DGX Spark (GB10) cluster

Recipe: GLM-5.2 (unpruned QuantTrio Int4-Int8Mix) at 200K ctx with MTP spec decode on a 4x NVIDIA DGX Spark (GB10) cluster

Serve GLM-5.2 469B (REAP-pruned, NVFP4) across 3× NVIDIA DGX Spark with vLLM pipeline parallelism — 256K context, production-ready config and patches

A lightweight cli for running single-purpose AI agents. Define focused agents in TOML, trigger them from anywhere; pipes, git hooks, cron, or the terminal.

Resumes generated using the GitHub informations

macOS cross compiler toolchains

A Rust HAL for the nRF family of devices

Warhammer 40,000: 9th Edition

High throughput transfer demo for iOS

Align videos/sound files timewise with help of their soundtracks

a small protobuf implementation in C

🥧 Savoury implementation of the QUIC transport protocol and HTTP/3

A library for performance traces from production.

Async-friendly QUIC implementation in Rust

Internet-Drafts that make up the base QUIC specification

Minimal implementation of the QUIC protocol

TLS 1.3 Specification

a fast, scalable, multi-language and extensible build system

A simple Go implementation of QUIC

Library that helps in instrumenting battery related system metrics.

An open-source C++ library developed and used at Facebook.

Stack trace symbolication library written in Rust