#cuda

[DEPRECATED] Moved to ROCm/rocm-libraries repo

Out-of-tree vLLM plugin and open format spec for NVFP4-CB / FP8-CB product-codebook weights — 2-6 bit-per-weight LLM quantization served on native Blackwell tensor cores.

Run AI models locally on your machine with node.js bindings for llama.cpp. Enforce a JSON schema on the model output on the generation level

Safe Rust wrapper for the NVIDIA Management Library

Fused CUDA kernels for LLM inference - RMSNorm / RoPE / SwiGLU

Compiler-grade sequential GPU compute. Workgroup-local stacks, queues, hashmaps, dominator trees, fixed-point dataflow. CUDA + WGPU + SPIR-V with bit-exact conformance gate. Rust.

ArWen: GPU-native WRF-ARW-class regional weather model, verified against WRF v4.6.1

GPU-accelerated semantic search for your docs and source code - hybrid dense + sparse RAG on a local Qdrant backend, served to Claude Code and other MCP clients. The search companion to vaultspec-core.

A simple service to publish NVIDIA GPU statistics from nvidia-smi to a MQTT broker and Home Assistent with Auto-Discovery.

Production-grade Rust/Python quantum error correction decoder. 25+ decoder families: MWPM Blossom, Union-Find, BP-OSD, LDPC/qLDPC, belief-matching, CUDA/GPU batch decoding, AutoDecoder 7-tier fallback. PyMatching/Stim/Sinter compatible. Ed25519 license verification. Benchmark evidence included.

Llama cpp + CapacitorJS support

A nvImageCodec library of GPU- and CPU- accelerated codecs featuring a unified interface