63 lines
1.1 KiB
Markdown
63 lines
1.1 KiB
Markdown
|
---
|
||
|
|
pipeline_tag: text-generation
|
||
|
|
language:
|
||
|
|
- en
|
||
|
|
tags:
|
||
|
|
- UnfilteredAI
|
||
|
|
- DAN
|
||
|
|
- NSFW
|
||
|
|
- Unfiltered
|
||
|
|
- Toxic-AI
|
||
|
|
- not-for-all-audiences
|
||
|
|
- Llama3
|
||
|
|
- mlx
|
||
|
|
library_name: mlx
|
||
|
|
base_model: UnfilteredAI/DAN-L3-R1-8B
|
||
|
|
license: apache-2.0
|
||
|
|
---
|
||
|
|
|
||
|
|
# Vlor999/UnfilteredAI-DAN-L3-R1-8B
|
||
|
|
|
||
|
|
This model [Vlor999/UnfilteredAI-DAN-L3-R1-8B](https://huggingface.co/Vlor999/UnfilteredAI-DAN-L3-R1-8B) was
|
||
|
|
converted to MLX format from [UnfilteredAI/DAN-L3-R1-8B](https://huggingface.co/UnfilteredAI/DAN-L3-R1-8B)
|
||
|
|
using mlx-lm version **0.29.1**.
|
||
|
|
|
||
|
|
## Use with mlx
|
||
|
|
|
||
|
|
With the **uv** version:
|
||
|
|
```bash
|
||
|
|
# Init the repository
|
||
|
|
uv init .
|
||
|
|
# Add the dependencies:
|
||
|
|
uv add mlx-lm
|
||
|
|
```
|
||
|
|
|
||
|
|
With the **pip** version:
|
||
|
|
```bash
|
||
|
|
pip install mlx-lm
|
||
|
|
```
|
||
|
|
|
||
|
|
Create a python file (ex: main.py)
|
||
|
|
```python
|
||
|
|
from mlx_lm import load, generate
|
||
|
|
|
||
|
|
model, tokenizer = load("Vlor999/UnfilteredAI-DAN-L3-R1-8B")
|
||
|
|
|
||
|
|
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)
|
||
|
|
```
|
||
|
|
|
||
|
|
To run it:
|
||
|
|
```bash
|
||
|
|
# uv version
|
||
|
|
uv run python main.py
|
||
|
|
```
|
||
|
|
|