Instructions to use deepseek-ai/deepseek-coder-33b-instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use deepseek-ai/deepseek-coder-33b-instruct with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="deepseek-ai/deepseek-coder-33b-instruct")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("deepseek-ai/deepseek-coder-33b-instruct")
model = AutoModelForCausalLM.from_pretrained("deepseek-ai/deepseek-coder-33b-instruct")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use deepseek-ai/deepseek-coder-33b-instruct with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "deepseek-ai/deepseek-coder-33b-instruct"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "deepseek-ai/deepseek-coder-33b-instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/deepseek-ai/deepseek-coder-33b-instruct

SGLang

How to use deepseek-ai/deepseek-coder-33b-instruct with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "deepseek-ai/deepseek-coder-33b-instruct" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "deepseek-ai/deepseek-coder-33b-instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "deepseek-ai/deepseek-coder-33b-instruct" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "deepseek-ai/deepseek-coder-33b-instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use deepseek-ai/deepseek-coder-33b-instruct with Docker Model Runner:
```
docker model run hf.co/deepseek-ai/deepseek-coder-33b-instruct
```

This open source version performs differently from the official trial version. Are these two different versions?

#12

by ZHUYONGJUN - opened Nov 28, 2023

Discussion

ZHUYONGJUN

Nov 28, 2023

•

edited Nov 28, 2023

The performance of the same prompt is inconsistent between the open source version and the trial version of https://chat.deepseek.com/. The official website trial version performs much better, while the open source version performs very poorly. Are these two different versions?

ZHUYONGJUN changed discussion title from 这个开源版本与官网试用版本表现效果不一样，这是两个不同的版本吗？ to This open source version performs differently from the official trial version. Are these two different versions? Nov 28, 2023

geepytee

Dec 13, 2023

reduce the Repetition penalty to 1, the code will be much better, and closely resemble what is generated on the website. (tested multiple times with pong and snake)

luofuli

Feb 4, 2024

The official website https://chat.deepseek.com/ has recently been updated in January with an SFT-optimized model, so it will be somewhat better than the open-source version. This model might be open-sourced in the future.

luofuli changed discussion status to closed Feb 4, 2024

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment