itsVentie/nano-guard-gguf

Nano-Guard Quantized & Compiled Artifacts

This repository provides optimized binary weights (ONNX and GGUF execution formats) for itsVentie/nano-guard-classifier-v1.

Source Code & Engine Integration: itsVentie/nano-guard on GitHub

File Inventory

Filename	Format	Target Runtime	Purpose
model.onnx	ONNX (Oset 14)	ort (Rust) / ONNX Runtime	Embedded execution in nano-guard-core
model-q4_k_m.gguf	GGUF (4-bit quantization)	llama.cpp / Local Executables	Low-memory execution environments
tokenizer.json	JSON	tokenizers crate	Fast Rust-native tokenization

ONNX Execution via Rust (ort)

use ort::{GraphOptimizationLevel, Session};
use tokenizers::Tokenizer;

let tokenizer = Tokenizer::from_file("tokenizer.json")?;
let session = Session::builder()?
    .with_optimization_level(GraphOptimizationLevel::Level3)?
    .commit_from_file("model.onnx")?;