Instructions to use deepseek-ai/deepseek-llm-7b-base with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use deepseek-ai/deepseek-llm-7b-base with Transformers:

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

pipe = pipeline("text-generation", model="deepseek-ai/deepseek-llm-7b-base")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("deepseek-ai/deepseek-llm-7b-base")
model = AutoModelForCausalLM.from_pretrained("deepseek-ai/deepseek-llm-7b-base")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use deepseek-ai/deepseek-llm-7b-base with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "deepseek-ai/deepseek-llm-7b-base"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "deepseek-ai/deepseek-llm-7b-base",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/deepseek-ai/deepseek-llm-7b-base

SGLang

How to use deepseek-ai/deepseek-llm-7b-base 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-llm-7b-base" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "deepseek-ai/deepseek-llm-7b-base",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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-llm-7b-base" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "deepseek-ai/deepseek-llm-7b-base",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use deepseek-ai/deepseek-llm-7b-base with Docker Model Runner:
```
docker model run hf.co/deepseek-ai/deepseek-llm-7b-base
```

LlamaTokenizer and LlamaTokenizerFast

by chim2000 - opened Apr 19, 2024

Discussion

chim2000

Apr 19, 2024

I used mmlu as eval dataset for deepseek-7b base, but the result acc = 0. After switching to LlamaTokenizer as tokenizer, acc comes to 0.249, which proves nothing goes wrong with the inference codes. Could you explain the difference between LlamaTokenizer and LlamaTokenizerFast, or please provide me with your prompts?

sanderland

May 16, 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