Instructions to use mlx-community/DeepSeek-R1-0528-5bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use mlx-community/DeepSeek-R1-0528-5bit with MLX:
# Make sure mlx-lm is installed # pip install --upgrade mlx-lm # Generate text with mlx-lm from mlx_lm import load, generate model, tokenizer = load("mlx-community/DeepSeek-R1-0528-5bit") prompt = "Write a story about Einstein" messages = [{"role": "user", "content": prompt}] prompt = tokenizer.apply_chat_template( messages, add_generation_prompt=True ) text = generate(model, tokenizer, prompt=prompt, verbose=True) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- MLX LM
How to use mlx-community/DeepSeek-R1-0528-5bit with MLX LM:
Generate or start a chat session
# Install MLX LM uv tool install mlx-lm # Interactive chat REPL mlx_lm.chat --model "mlx-community/DeepSeek-R1-0528-5bit"
Run an OpenAI-compatible server
# Install MLX LM uv tool install mlx-lm # Start the server mlx_lm.server --model "mlx-community/DeepSeek-R1-0528-5bit" # Calling the OpenAI-compatible server with curl curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "mlx-community/DeepSeek-R1-0528-5bit", "messages": [ {"role": "user", "content": "Hello"} ] }'
Add files using upload-large-folder tool
Browse files- README.md +28 -0
- config.json +0 -0
README.md
CHANGED
|
@@ -6,3 +6,31 @@ tags:
|
|
| 6 |
- mlx
|
| 7 |
pipeline_tag: text-generation
|
| 8 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 6 |
- mlx
|
| 7 |
pipeline_tag: text-generation
|
| 8 |
---
|
| 9 |
+
|
| 10 |
+
# mlx-community/DeepSeek-R1-0528-5bit
|
| 11 |
+
|
| 12 |
+
This model [mlx-community/DeepSeek-R1-0528-5bit](https://huggingface.co/mlx-community/DeepSeek-R1-0528-5bit) was
|
| 13 |
+
converted to MLX format from [deepseek-ai/DeepSeek-R1-0528](https://huggingface.co/deepseek-ai/DeepSeek-R1-0528)
|
| 14 |
+
using mlx-lm version **0.25.3**.
|
| 15 |
+
|
| 16 |
+
## Use with mlx
|
| 17 |
+
|
| 18 |
+
```bash
|
| 19 |
+
pip install mlx-lm
|
| 20 |
+
```
|
| 21 |
+
|
| 22 |
+
```python
|
| 23 |
+
from mlx_lm import load, generate
|
| 24 |
+
|
| 25 |
+
model, tokenizer = load("mlx-community/DeepSeek-R1-0528-5bit")
|
| 26 |
+
|
| 27 |
+
prompt = "hello"
|
| 28 |
+
|
| 29 |
+
if tokenizer.chat_template is not None:
|
| 30 |
+
messages = [{"role": "user", "content": prompt}]
|
| 31 |
+
prompt = tokenizer.apply_chat_template(
|
| 32 |
+
messages, add_generation_prompt=True
|
| 33 |
+
)
|
| 34 |
+
|
| 35 |
+
response = generate(model, tokenizer, prompt=prompt, verbose=True)
|
| 36 |
+
```
|
config.json
CHANGED
|
The diff for this file is too large to render.
See raw diff
|
|
|