A Rust client library for real-time updates from the Tensor NFT marketplace and its aggregated marketplaces, using GraphQL subscriptions.
A client for receiving updates from Tensor Trade Marketplace over WebSocket using GraphQL subscriptions.
Transport protocols and zero-copy data exchange for IPFRS distributed system
Core BitNet implementation with fundamental data structures and algorithms
Distributed runtime with QUIC transport, stream-multiplexed messaging, and built-in collectives
A tensor library with GPU support
Tensor decompositions: CP-ALS, Tucker-HOOI, TT-SVD
Attestation-bound encrypted tensor transport for confidential ML inference
AXON v1.5.1 — first crates.io publication of the AXON language full-stack runtime. Lexer/parser/type-checker/IR generator (re-exported from axon-frontend) plus the native Rust runtime: typed channels (TypedEventBus with QoS×5, π-calculus mobility, capability extrusion via shield D8 — Fase 13.f.2), Free Monad CPS handlers (Fase 2), lease kernel + reconcile loop (Fase 3+5), Epistemic Security Kernel (ESK Fase 6), Trust Types + ReplayLog (Fase 11.a+11.c), Stateful PEM over WebSocket (Fase 11.d), Ontological Tool Synthesis (Fase 11.e), Mobile Typed Channels (Fase 13). Crate publishes as `axon-lang` to mirror the Python PyPI package; library import remains `use axon::*` so existing call sites keep working unchanged.
EML operator: all elementary functions from exp(x) - ln(y)
Temporal tensor compression with tiered quantization for RuVector
Asynchronous Batched Inference Platform
A reimplementation of TensorFlow for ruby. This is a ground up implementation with no dependency on TensorFlow. Effort has been made to make the programming style as near to TensorFlow as possible, comes with a pure ruby evaluator by default as well with support for an opencl evaluator.
OpenCL evaluator for tensor_stream