Releases

Megatron-LM-style tensor-parallel primitives (Column / Row Parallel Linear) for the baracuda CUDA stack. Pure-composition crate — local GEMM via baracuda-cublas + cross-rank collectives via baracuda-nccl. No new CUDA kernels. NEW in Phase 57; deliberate scope expansion (distributed-training-framework-adjacent). Off-by-default in baracuda-kernels via the `megatron_tp` cargo feature so non-distributed consumers don't pay the dep surface cost. Algorithmic reference: Shoeybi et al. arXiv:1909.08053 (NVIDIA Megatron-LM, Apache-2.0).

Idiomatic Rust wrappers for the NVIDIA CUDA stack (Driver API, Runtime API, NVRTC, cuBLAS, cuDNN, NCCL, NVML, ...). Umbrella crate.

Safe Rust wrappers for the CUDA Runtime API (devices, streams, events, managed memory, kernel launch via the library API).

Build + raw FFI bindings to baracuda's clean-fork of Hiroyuki Ootomo's ozIMMU — the Ozaki-scheme FP64 GEMM library that synthesizes a DGEMM from S² int8 tensor-core matmuls. Phase 44b internalized the upstream sources under `cuda/` (no more `vendor/` subdir; cutf submodule eliminated). Linked statically into the baracuda CUDA stack; consumed by the safe wrapper crate `baracuda-ozimmu`. MIT-licensed (original ozIMMU MIT — see `ATTRIBUTION.md`).

Safe Rust wrappers for NVIDIA NVRTC (compile CUDA C++ to PTX at runtime).

The Arora behavior tree: Groot-compatible trees ticking Arora modules.

Fish Audio Drama 3 text-to-speech in Rust (CUDA or Apple Silicon MLX)

Safe Rust wrappers for NVIDIA NPP (Performance Primitives). Core + signal subset at v0.1.

Safe Rust wrappers for NVIDIA cuVS (GPU vector search / approximate nearest neighbours).

Safe Rust wrappers for NVIDIA cuSOLVER (dense LU factorization at v0.1).

tailscale direct (disco) UDP underlay transport

Safe Rust wrappers for NVIDIA cuFFT.

userspace netstack built on smoltcp (socket wrappers)

Fast code index for C# and TypeScript codebases -- map, find, refs, impact from the CLI.

Build + raw FFI bindings to baracuda's port of NVIDIA TransformerEngine's FP8 cast/transpose + delayed-scaling recipe primitives. Cast/recipe subset only — `normalization` / `fused_rope` / `fused_attn` / `fused_softmax` / `activation` / `gemm` deliberately skipped (overlap existing baracuda Phase 3/5/14/17/30/31/36/41/42). NO cuDNN dep (recipe + cast paths don't need it; `fused_attn` would, and we skip it); NO pybind11 (the safe wrapper lives in `baracuda-transformer-engine` and exposes a raw C ABI defined in `csrc/baracuda_te_shim.cu`). Apache-2.0 per upstream — see `ATTRIBUTION.md`.

Raw FFI bindings and dynamic loader for NVIDIA nvJPEG.

Safe Rust wrappers for the CUDA Driver API (devices, contexts, streams, events, memory, kernels, graphs).

Raw FFI bindings and dynamic loader for NVIDIA cuVS (GPU vector search / ANN).

Raw C-ABI FFI surface for the vendored FlashInfer inference kernels (paged-KV decode, paged-KV append, cascade state merge, sort-free sampling). The launcher shims + vendored FlashInfer sources are compiled by `baracuda-kernels-sys`; this crate is a thin re-export facade so the FlashInfer C-ABI is reachable under its own crate name. Apache-2.0 (FlashInfer upstream) — see `crates/baracuda-kernels-sys/vendor/flashinfer/`.

Safe Rust wrappers for CUPTI. Scaffolding at v0.1.

Compiled bespoke .cu kernel template instantiations for the baracuda ML kernel facade plus C-ABI FFI facades for the library-backed plans (cuDNN conv/pool, cuSOLVER linalg, cuFFT/cuRAND, CUTLASS GEMM re-export). Hosts curated CUDA kernel sources (int8/FP8/int4/bin GEMM RRR, elementwise, reduce, norm, attention, …), builds them via baracuda-forge, exposes extern "C" entry points for the safe baracuda-kernels crate. CUTLASS template kernels live in the sibling baracuda-cutlass-kernels-sys crate and are re-exported here under the unified baracuda_kernels_gemm_* namespace.

Raw FFI bindings and dynamic loader for the CUDA Driver and Runtime APIs (libcuda / libcudart).

Compiled CUTLASS template instantiations for the baracuda ecosystem. Hosts curated .cu kernel sources, builds them via baracuda-forge, exposes extern "C" entry points for the safe baracuda-cutlass crate.

Client-side wire helpers for the capability Trust Task families (governance/capability/*, git-trust/*): document builders, envelope parsing, reply classification. Shared by capability producers (community services) and management UIs so the two cannot drift.

DIDComm v2.1 transport binding for the Trust Tasks framework, built on affinidi-messaging-didcomm. Provides pack/unpack helpers and a TransportHandler that maps the verified envelope sender into the framework's §4.8.1 precedence.

ToIP Trust Spanning Protocol (TSP) transport binding for the Trust Tasks framework, built on affinidi-tsp. Provides pack/unpack helpers and a TransportHandler that maps the authenticated VID sender into the framework's §4.8.1 precedence.

One-dependency front door to the Trust Tasks framework: re-exports trust-tasks-rs plus the transport, proof, ceremony and capability crates behind Cargo features.

Pluggable ProofVerifier implementations for the Trust Tasks framework. Backends are gated by Cargo features; ships with an Affinidi Data Integrity backend (EdDSA suites) by default.

HTTPS transport binding for the Trust Tasks framework — typed client + axum-based server with bearer-auth identity, suitable for demos, mockups, and end-to-end testing.

Reference Rust library for the Trust Tasks framework — transport-agnostic, JSON-based descriptions of verifiable work between parties.