Releases

Rust SDK for programmatic control of the GitHub Copilot CLI via JSON-RPC.

Local-first data pipeline engine for analytical workflows

API bindings for the Nominal platform

Oxen server is a fast data version control backend, supporting local disk and S3. Self host your repositories on your own storage, or use the hosted platform on Oxen.ai. Stores, syncs, and serves versioned datasets, model checkpoints, game assets, studio media, and any large data. Use the oxen CLI to push and pull from the oxen server.

Oxen is a fast data version control system, built with machine learning training data in mind. Designed to handle terabytes of data with ease, using a workflow similar to git. Version both structured and unstructured data of any modality: text, images, video, audio, CSV, Parquet, JSONL, model checkpoints, and more. liboxen is the embeddable core library behind the oxen CLI and server, which power fine tuning and inference pipelines for multimodal LLMs, image models, and video models on Oxen.ai.

Opinionated Arora runtime: an engine pre-wired with the native behavior-tree core nodes and the Semio backend.

The Arora engine: loads and executes Arora modules (wasm and native hosts).

Native Rust plugin for Bun applications using bun:ffi.

Mbedtls based crypto provider for rustls.

Unified ML op facade for the baracuda CUDA ecosystem. Exposes every primitive an ML framework would expect (union of PyTorch torch.* + nn.functional and JAX lax.* / numpy ops) through a single Plan-based Rust surface, internally dispatching to baracuda-cutlass, the baracuda-* NVIDIA-library wrappers, or bespoke baracuda-kernels-sys kernels.

Safe Rust wrapper for compiled CUTLASS kernels: plan-based GEMM and grouped GEMM with caller-supplied workspace, typed device-buffer arguments, and capture-safe launch.

Safe Rust wrapper for baracuda's clean-fork of Hiroyuki Ootomo's ozIMMU — Ozaki-scheme FP64 GEMM that synthesizes a DGEMM from S^2 int8 tensor-core matmuls. Provides an RAII handle + drop-in `dgemm` shim suitable for the `BackendKind::Ozaki` path of `baracuda-kernels`'s `GemmPlan`. Opt-in (NOT bit-equivalent to native DGEMM); the default FP64 path stays on CUTLASS / cuBLAS.

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).

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`).

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`.

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

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.

Fluid Attacks design system, GPUI variant: desktop widgets over the shared tokens

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.

Encode and decode HEIF/HEIC images with x265 and libde265, via statically-linked libheif.

Encode and decode WebP images via statically-linked libwebp.

Encode and decode AVIF images with SVT-AV1 and dav1d, via statically-linked libavif.

Standalone TUI infoview for Lean 4 theorem prover

Opinionated Arora runtime: an engine pre-wired with the native behavior-tree core nodes and the Semio backend.

The Arora engine: loads and executes Arora modules (wasm and native hosts).

An actor-first distributed database

Standalone regional durable-actors control plane, host, and durability runtime

APIs for creating shielded Zcash light clients

An SQLite-based Zcash light client