README.md

---
base_model: DiscoResearch/DiscoLM_German_7b_v1
tags:
- Mistral
- finetune
- chatml
- DPO
- German
- Deutsch
- synthetic data
- mlx
license: apache-2.0
language:
- de
- en
model-index:
- name: DiscoLM_German_7b_v1
  results: []
---

# Purpose

mlx-converted version of [DiscoResearch/DiscoLM_German_7b_v1](https://huggingface.co/DiscoResearch/DiscoLM_German_7b_v1) with convincing german capacities. 
Done mainly for ollama import, which was not possible with the base model (unless you pick a gguf)

# mlx-community/DiscoLM_German_7b_v1-mlx

The Model [mlx-community/DiscoLM_German_7b_v1-mlx](https://huggingface.co/mlx-community/DiscoLM_German_7b_v1-mlx) was
converted to MLX format from [DiscoResearch/DiscoLM_German_7b_v1](https://huggingface.co/DiscoResearch/DiscoLM_German_7b_v1)
using mlx-lm version **0.21.4**.

## Use with mlx

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

```python
from mlx_lm import load, generate

model, tokenizer = load("jachermann/DiscoLM_German_7b_v1-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
    )

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