----> We don't Own this model this Is prepared For Hackthaon
📂 Model Organization (Curated by SKT AI Labs)
To make it easier for Indian developers to use Google's power, we have categorized the models as follows:
1. 💻 SFT-CODER-GGUF
Assembled for software engineering tasks.
- Base Models: Google Gemma-4 Coding Variants.
- Format: GGUF (Optimized for local/edge inference).
2. 🧠 SFT-REASONING
Assembled for logic, math, and deep reasoning.
- Base Models: Google Gemma-4 Instruct/Reasoning Variants.
- Focus: High-parameter logic processing.
3. 🏗️ SFT-MODEL
Full weights/Safetensors for research and fine-tuning.
🚀 Purpose for the Hackathon
The goal of this repository is to showcase how Google's Gemma-4 can be utilized to build "Sovereign Intelligence" solutions. By organizing these models into specialized tiers, we demonstrate a scalable architecture for using Google's technology in diverse real-world scenarios.
⚖️ Licensing
These models are released under the Google Gemma Terms of Use. Please ensure you comply with Google's acceptable use policy when utilizing these weights.
