AnkitBirGurung/gemma-4-12b-platinumSFT-gguf overview
Using Neutral Dataset, I've enhanced the thinking of the Model. Please use 0.45 as temperature to use this Model. I've Used a Generic Heretic Model + SFT on Bu…
Runs locally from ~6.87 GB disk (8 GB VRAM class GPUs with llama.cpp / guIDE).
Repository Files & Downloads
| File | Type | Quantization | Size | Link |
|---|---|---|---|---|
| gemma-4-12B-it-heretic-v1.Q4_K_M.gguf | GGUF | GGUF | 6.87 GB | Download |
Model Details
| Model ID | AnkitBirGurung/gemma-4-12b-platinumSFT-gguf |
|---|---|
| Author | AnkitBirGurung |
| Pipeline | — |
| License | apache-2.0 |
| Base model | igorls/gemma-4-12B-it-heretic-GGUF |
| Last modified | 2026-07-07T15:26:20.000Z |
Model README
---
license: apache-2.0
language:
- en
base_model:
- igorls/gemma-4-12B-it-heretic-GGUF
tags:
- '#SuperFineTuned'
---
Using Neutral Dataset, I've enhanced the thinking of the Model. Please use 0.45 as temperature to use this Model.
I've Used a Generic Heretic Model + SFT on Business, Psychology, Genius Way of Thinking. If you Give it a LAW RAG. It is better able to answer you questions.
The Dataset was Created based on General Business Talk. It was Enhanced Using OpenRouter. It uses Fable, Opus, Deepseek, Mistral, LLama, Qwen: all combined
To Prepare the Dataset.
Some Were Customized to Train the Model in Platinum Level 30 Dataset to make it SFT.
Run AnkitBirGurung/gemma-4-12b-platinumSFT-gguf with guIDE
Download guIDE — the AI-native code editor with local LLM inference and 69 built-in tools.
Source: Hugging Face · Compare models