Releases

Wire-protocol constants, codecs and the circuit graph shared by gnitz clients and the gnitz engine

Heddle's CLI verb catalog, schemas, help, and recovery contract.

A component of the servo web-engine.

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

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.

This library provides the definition of the enclave image format (EIF) file used in AWS Nitro Enclaves.

The tari full base node implementation

In-process full Mesh LLM node embedding API

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`.

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`).

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.

Service for Verifiable Trust Agents operating in Verifiable Trust Communities

Host system inspection and update helpers for mesh-llm

Native runtime manifest discovery, download, installation, and cache management for Mesh LLM

Deadline-safe native serving-plugin loader for Mesh Skippy

Embedded Skippy staged runtime server

Bitcoin Commons BLVM: Direct mathematical implementation of Bitcoin consensus rules from the Orange Paper

Iteration-level scheduler for concurrent staged serving with unified KV

Rust runtime layer for Skippy staged model execution

VTA persona store — the holder's own identity attributes, the profiles that project over them, the bindings that assign a profile to a persona DID, and the contacts peers disclose

VTA TEE bootstrap — Nitro/SEV-SNP attestation providers, KMS attest/decrypt + storage-key derivation, the anchor MAC, and first-boot DID autogen

luma flash attn

WebVH hosting infrastructure for the VTA — the local did:webvh DID-record and hosting-server store

A GPU renderer for Vello with CPU-side preprocessing.

VTA holder credential vault — storage, query, receive/verify, present, and status refresh for credentials a holder stores on a VTA

Background TTL sweepers for the VTA's core keyspaces — ACL grant expiry, pending-consent expiry, and soft-deleted-vault purge

VTA policy subsystem — the regorus (Rego) engine, the default policy bundle, consent model, and decision evaluators

Fast, lightweight and schema-less search backend.

VTA backup/restore subsystem — encrypted full-state export/import, the sealed backup-bundle store, and its TTL sweeper

VTA key management — master-seed storage, BIP-32 key derivation, key wrapping, and the seed-store backend selection