Instructions to use DMindAI/DMind-1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use DMindAI/DMind-1 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="DMindAI/DMind-1") messages = [ {"role": "user", "content": "Who are you?"}, ] pipe(messages)# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("DMindAI/DMind-1", device_map="auto") - Inference
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use DMindAI/DMind-1 with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "DMindAI/DMind-1" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "DMindAI/DMind-1", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/DMindAI/DMind-1
- SGLang
How to use DMindAI/DMind-1 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 "DMindAI/DMind-1" \ --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": "DMindAI/DMind-1", "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 "DMindAI/DMind-1" \ --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": "DMindAI/DMind-1", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }' - Docker Model Runner
How to use DMindAI/DMind-1 with Docker Model Runner:
docker model run hf.co/DMindAI/DMind-1
Update handler.py
Browse files- handler.py +16 -17
handler.py
CHANGED
|
@@ -43,22 +43,21 @@ class EndpointHandler:
|
|
| 43 |
)
|
| 44 |
|
| 45 |
def __call__(self, data: Dict[str, Any]) -> Dict[str, str]:
|
| 46 |
-
"""
|
| 47 |
-
data 格式:
|
| 48 |
-
{
|
| 49 |
-
"inputs": "your prompt here"
|
| 50 |
-
}
|
| 51 |
-
"""
|
| 52 |
prompt = data["inputs"]
|
| 53 |
-
|
| 54 |
-
#
|
| 55 |
-
|
| 56 |
-
|
| 57 |
-
|
|
|
|
| 58 |
with torch.inference_mode():
|
| 59 |
-
output_ids = self.model.generate(
|
| 60 |
-
|
| 61 |
-
|
| 62 |
-
|
| 63 |
-
|
| 64 |
-
return {
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 43 |
)
|
| 44 |
|
| 45 |
def __call__(self, data: Dict[str, Any]) -> Dict[str, str]:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 46 |
prompt = data["inputs"]
|
| 47 |
+
|
| 48 |
+
# ① 自动抓 embedding 所在 GPU
|
| 49 |
+
first_device = next(self.model.parameters()).device
|
| 50 |
+
inputs = self.tokenizer(prompt, return_tensors="pt").to(first_device)
|
| 51 |
+
|
| 52 |
+
# ② 生成(其余逻辑不变)
|
| 53 |
with torch.inference_mode():
|
| 54 |
+
output_ids = self.model.generate(
|
| 55 |
+
**inputs,
|
| 56 |
+
**self.generation_kwargs,
|
| 57 |
+
)
|
| 58 |
+
|
| 59 |
+
return {
|
| 60 |
+
"generated_text": self.tokenizer.decode(
|
| 61 |
+
output_ids[0], skip_special_tokens=True
|
| 62 |
+
)
|
| 63 |
+
}
|