Crate kimi_fann_core 

Kimi-FANN Core: Neural Inference Engine

Real neural network inference engine using ruv-FANN for micro-expert processing. This crate provides WebAssembly-compatible neural network inference with actual AI processing capabilities for Kimi-K2 micro-expert architecture. Enhanced with Synaptic Market integration for distributed compute.

Modules

enhanced_router

Enhanced routing system with market integration

optimized_features

Optimized Feature Extraction for Neural Inference

Structs

ExpertConfig

Configuration for creating a micro-expert

ExpertRouter

Expert router with intelligent request distribution and consensus

KimiRuntime

Main runtime for Kimi-FANN with neural processing

MicroExpert

A micro-expert neural network with real AI processing

MockNeuralNetwork

Mock neural network for demonstration and optimization testing

NetworkStats

Network statistics for distributed processing

NeuralConfig

Neural network configuration for micro-experts

NeuralWeights

Neural network weights and biases

ProcessingConfig

Processing configuration with neural parameters

TokenEmbedding

Token embedding for neural processing

Enums

ExpertDomain

Expert domain enumeration with neural specialization

Constants

VERSION

Version of the Kimi-FANN Core library

Functions

init

Initialize the WASM module with neural network setup