Update README.md

README.md

@@ -10,6 +10,40 @@ widget:
 
 This model is for debugging. It is randomly initialized with the config from [deepseek-ai/DeepSeek-V3](https://huggingface.co/deepseek-ai/DeepSeek-V3) but is of smaller size. 
 
+Usage:
+```python
+import torch
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+model_id = "yujiepan/deepseek-v3-tiny-random"
+device = torch.device("cuda")
+
+tokenizer = AutoTokenizer.from_pretrained(model_id)
+model = AutoModelForCausalLM.from_pretrained(
+    model_id, trust_remote_code=True,
+).eval().to(device)
+
+prompt = 'Hello!'
+messages = [
+    {"role": "system", "content": "You are a helpful assistant."},
+    {"role": "user", "content": prompt}
+]
+
+inputs = tokenizer.apply_chat_template(
+    messages, tokenize=True, add_generation_prompt=True, return_tensors="pt"
+).to(device)
+
+with torch.inference_mode():
+    outputs = model.generate(
+        inputs,
+        max_new_tokens=16,
+        do_sample=False,
+        use_cache=True,
+    )
+string = tokenizer.decode(outputs[0])
+print(string)
+```
+
 Codes:
 ```python
 import os