Instructions to use deepseek-ai/DeepSeek-V3.2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use deepseek-ai/DeepSeek-V3.2 with Transformers:

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

pipe = pipeline("text-generation", model="deepseek-ai/DeepSeek-V3.2")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("deepseek-ai/DeepSeek-V3.2", dtype="auto")

Inference
HuggingChat
Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use deepseek-ai/DeepSeek-V3.2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "deepseek-ai/DeepSeek-V3.2"
# 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-V3.2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/deepseek-ai/DeepSeek-V3.2

SGLang

How to use deepseek-ai/DeepSeek-V3.2 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-V3.2" \
    --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-V3.2",
		"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-V3.2" \
        --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-V3.2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use deepseek-ai/DeepSeek-V3.2 with Docker Model Runner:
```
docker model run hf.co/deepseek-ai/DeepSeek-V3.2
```

Running the model with a dense attention

#35

by sszymczyk - opened Jan 2

Discussion

sszymczyk

Jan 2

Since the model is not yet supported in llama.cpp I did an experiment and ran DeepSeek V3.2 with dense attention by disabling lightning indexer. The model seems to perform just fine - at least based on my limited testing. Do you think the model performance (by this I mean intelligence, not speed) will be affected by doing this?

Fujikre

Jan 3

Yes

sszymczyk

Jan 6

@Fujikre
I tried quite hard to find a difference in the model intelligence. I tested Q4_K_M quantized DeepSeek V3.2 GGUF with disabled lightning indexer in regular llama.cpp build by running my lineage-bench logical reasoning benchmark. The result (numbers are mean accuracy for given difficulty, there are 4 difficulty levels, for each difficulty there are 40 quizzes):

Nr	model_name	lineage	lineage-8	lineage-64	lineage-128	lineage-192
1	deepseek/deepseek-v3.2	0.988	1.000	1.000	1.000	0.950

So DeepSeek V3.2 with dense attention solved correctly almost all quizzes (there were 160 quizzes overall), it got wrong only 2 most difficult quizzes. When I tested the original model via API the result was:

Nr	model_name	lineage	lineage-8	lineage-64	lineage-128	lineage-192
1	deepseek/deepseek-v3.2	0.956	1.000	1.000	0.975	0.850

So it looks like the model with dense attention performed even a bit better than the sparse attention one.
Or perhaps you meant that the model intelligence will be affected, but positively?

bibproj

Jan 6

I tested the experimental Q4_K_M quantized DeepSeek V3.2 GGUF (dense attention) on translating documents. The translations are similar to the ones I get from the MLX quants that use sparse attention, and similar to the API results.

sszymczyk

Mar 10

I revisited this after some time and found clear difference in reasoning performance in more complex tasks. In short using dense attention makes the model a bit dumber. Details here: https://www.reddit.com/r/LocalLLaMA/comments/1rq8otd/running_deepseek_v32_with_dense_attention_like_in/

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