gb200

TokenSpeed is a speed-of-light LLM inference engine.

A 3D visualizaion of LLM training and inferencing on modern cluster

An empirical study of benchmarking LLM inference with KV cache offloading using vLLM and LMCache on NVIDIA GB200 with high-bandwidth NVLink-C2C .

Prebuilt spconv v2.3.8 wheels for CUDA 12.8 / 13.0 with native Blackwell (RTX 50-series, sm_120) kernels — for default PyPI torch (cu130) or torch +cu128

Software-in-the-loop behavioral model of a GPU cluster modelled on NVIDIA GB200 NVL72, scaled down to two compute trays, on one Linux machine. Real Slurm or Kubernetes, PyTorch and NCCL run against emulated GPUs (CUDA, NVML, nvidia-smi), NVLink/NVSwitch partitions, Redfish BMCs and InfiniBand.

Two NVIDIA DGX SuperPOD scalable units modeled in NetBox — air-cooled DGX H100 and liquid-cooled GB200 NVL72 — rebuilt from a YAML spec by an idempotent Python generator, verified in CI. Paper model, not affiliated with NVIDIA.

A plain-spoken, interactive course on NVIDIA's Blackwell architecture — the chips, engines, and rack-scale systems built for the age of AI reasoning.

Sovereign Sub-Millisecond Speculative Expert Dispatcher, Waterfill Affinity Load-Balancer & All-to-All Overlap Engine for DeepSeek R1 & NVIDIA Blackwell

Sovereign 32x KV-Cache Transform Coding (KVTC), Hierarchical Context Memory Storage (CMX) & Cross-Agent Attention Deduplication Engine