---
tags:
- gguf
- llama.cpp
- unsloth
- vision-language-model
license: apache-2.0
datasets:
- kintsugicollective/atlas-dataset-v8-final
language:
- en
base_model:
- google/gemma-4-E4B-it
---

# atlas-gemma4-e4b : GGUF

- This variant of Atlas is purely SFT'ed through Unsloth Studio. I wanted to trial Unsloth Studio and also a smaller parameter model for Atlas local deployment.
- This model has had NO refusal removal, only SFT using my bespoke data set. 


This model was finetuned and converted to GGUF format using [Unsloth](https://github.com/unslothai/unsloth).

**Example usage**:
- For text only LLMs:    `llama-cli -hf senaro/atlas-gemma4-e4b --jinja`
- For multimodal models: `llama-mtmd-cli -hf senaro/atlas-gemma4-e4b --jinja`

## Available Model files:
- `gemma-4-e4b-it.Q4_K_M.gguf`
- `gemma-4-e4b-it.BF16-mmproj.gguf`

## ⚠️ Ollama Note for Vision Models
**Important:** Ollama currently does not support separate mmproj files for vision models.

To create an Ollama model from this vision model:
1. Place the `Modelfile` in the same directory as the finetuned bf16 merged model
3. Run: `ollama create model_name -f ./Modelfile`
   (Replace `model_name` with your desired name)

This will create a unified bf16 model that Ollama can use.
This was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth)
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)