---
library_name: mlx
base_model:
- DavidAU/Gemma-3-27b-it-Uncensored-HERETIC-Gemini-Deep-Reasoning
datasets:
- TeichAI/gemini-3-pro-preview-high-reasoning-250x
language:
- en
- fr
- de
- es
- it
- pt
- ru
- zh
- ja
tags:
- reasoning
- thinking
- gemma3
- deep thinking
- finetune
- creative
- creative writing
- fiction writing
- plot generation
- sub-plot generation
- story generation
- scene continue
- storytelling
- fiction story
- science fiction
- romance
- all genres
- story
- writing
- vivid prose
- vivid writing
- fiction
- roleplaying
- bfloat16
- swearing
- rp
- unsloth
- context 128k
- mlx
pipeline_tag: text-generation
---

# Gemma-3-27b-it-HERETIC-Gemini-Deep-Reasoning-q8-mlx

Quantized model performance
```brainwave
q6  0.594,0.746,0.881,0.779,0.464,0.816,0.751
q8  0.596,0.748,0.881,0.779,0.458,0.819,0.751
```

Brainwaves for regular vs Heretic models
```brainwave
regular  0.590,0.742,0.883,0.781,0.458,0.822,0.751
heretic  0.596,0.748,0.881,0.779,0.458,0.819,0.751
```
Heretic ablation improved the model arc/arc_easy significantly, with minor drops in other places

Brainwaves for baseline vs Gemini trained model
```brainwave
gemma-3-27b-it-heretic
q8  0.557,0.711,0.868,0.533,0.452,0.706,0.695

Gemma-3-27b-it-HERETIC-Gemini-Deep-Reasoning
q8  0.596,0.748,0.881,0.779,0.458,0.819,0.751 
```
DavidAU's Gemini training was very successful, raising the model perfomance envelope on all metrics

-G

This model [Gemma-3-27b-it-HERETIC-Gemini-Deep-Reasoning-q8-mlx](https://huggingface.co/nightmedia/Gemma-3-27b-it-HERETIC-Gemini-Deep-Reasoning-q8-mlx) was
converted to MLX format from [DavidAU/Gemma-3-27b-it-Gemini-Deep-Reasoning](https://huggingface.co/DavidAU/Gemma-3-27b-it-Gemini-Deep-Reasoning)
using mlx-lm version **0.30.4**.

## Use with mlx

```bash
pip install mlx-lm
```

```python
from mlx_lm import load, generate

model, tokenizer = load("Gemma-3-27b-it-HERETIC-Gemini-Deep-Reasoning-q8-mlx")

prompt = "hello"

if tokenizer.chat_template is not None:
    messages = [{"role": "user", "content": prompt}]
    prompt = tokenizer.apply_chat_template(
        messages, add_generation_prompt=True, return_dict=False,
    )

response = generate(model, tokenizer, prompt=prompt, verbose=True)
```