⚛ Vigyan AI: Sovereign Quantized GGUF Suite [Research Preview]
This repository contains official GGUF quantizations of the Vigyan AI STEM Model Series engineered for 100% offline, air-gapped on-device deployment across consumer hardware, laptops, and edge devices.
📦 Available Checkpoints & Quantizations
| File | Quant Method | Target Hardware | Recommended Context |
|---|---|---|---|
vigyan-2b-reasoning-q4_k_m.gguf |
Q4_K_M (1.1 GB) | Mobile, WebGPU, 4GB RAM Laptops | 4,096 tokens |
vigyan-2b-reasoning-q8_0.gguf |
Q8_0 (1.9 GB) | High-precision on-device 8GB Laptops | 4,096 tokens |
🚀 Quickstart via llama.cpp
llama-cli -m vigyan-2b-reasoning-q8_0.gguf -t 4 --temp 0.2 --repeat-penalty 1.15 -p "<|user|>\nExplain Ohm's Law and RC time constant.\n<|assistant|>\n"
🔬 Research & Experimental Notice
These checkpoints represent intermediate research iterations developed by Vigyan AI to evaluate sovereign neuro-symbolic SLM execution combined with deterministic symbolic solvers (SymPy) and embedded C++ GraphRAG.