Instructions to use deepseek-ai/deepseek-vl2-tiny with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use deepseek-ai/deepseek-vl2-tiny with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-text-to-text", model="deepseek-ai/deepseek-vl2-tiny")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("deepseek-ai/deepseek-vl2-tiny", dtype="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps
- vLLM
How to use deepseek-ai/deepseek-vl2-tiny with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "deepseek-ai/deepseek-vl2-tiny" # 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-vl2-tiny", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/deepseek-ai/deepseek-vl2-tiny
- SGLang
How to use deepseek-ai/deepseek-vl2-tiny 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-vl2-tiny" \ --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-vl2-tiny", "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-vl2-tiny" \ --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-vl2-tiny", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use deepseek-ai/deepseek-vl2-tiny with Docker Model Runner:
docker model run hf.co/deepseek-ai/deepseek-vl2-tiny
Update README.md
#4
by Forgege - opened
README.md
CHANGED
|
@@ -34,6 +34,8 @@ DeepSeek-VL2-tiny is built on DeepSeekMoE-3B (total activated parameters are 1.0
|
|
| 34 |
On the basis of `Python >= 3.8` environment, install the necessary dependencies by running the following command:
|
| 35 |
|
| 36 |
```shell
|
|
|
|
|
|
|
| 37 |
pip install -e .
|
| 38 |
```
|
| 39 |
|
|
@@ -48,12 +50,12 @@ pip install -e .
|
|
| 48 |
import torch
|
| 49 |
from transformers import AutoModelForCausalLM
|
| 50 |
|
| 51 |
-
from
|
| 52 |
-
from
|
| 53 |
|
| 54 |
|
| 55 |
# specify the path to the model
|
| 56 |
-
model_path = "deepseek-ai/deepseek-vl2-
|
| 57 |
vl_chat_processor: DeepseekVLV2Processor = DeepseekVLV2Processor.from_pretrained(model_path)
|
| 58 |
tokenizer = vl_chat_processor.tokenizer
|
| 59 |
|
|
|
|
| 34 |
On the basis of `Python >= 3.8` environment, install the necessary dependencies by running the following command:
|
| 35 |
|
| 36 |
```shell
|
| 37 |
+
git clone https://github.com/deepseek-ai/DeepSeek-VL2
|
| 38 |
+
cd DeepSeek-VL2
|
| 39 |
pip install -e .
|
| 40 |
```
|
| 41 |
|
|
|
|
| 50 |
import torch
|
| 51 |
from transformers import AutoModelForCausalLM
|
| 52 |
|
| 53 |
+
from deepseek_vl2.models import DeepseekVLV2Processor, DeepseekVLV2ForCausalLM
|
| 54 |
+
from deepseek_vl2.utils.io import load_pil_images
|
| 55 |
|
| 56 |
|
| 57 |
# specify the path to the model
|
| 58 |
+
model_path = "deepseek-ai/deepseek-vl2-tiny"
|
| 59 |
vl_chat_processor: DeepseekVLV2Processor = DeepseekVLV2Processor.from_pretrained(model_path)
|
| 60 |
tokenizer = vl_chat_processor.tokenizer
|
| 61 |
|