Update README.md

README.md

@@ -67,7 +67,7 @@ Use the transformers backend for inference:
 ```python
 from transformers import AutoModelForCausalLM, AutoTokenizer
 
-MODEL_PATH = 'ZhipuAI/glm-4-9b-chat-1m-hf'
+MODEL_PATH = 'THUDM/glm-4-9b-chat-1m-hf'
 
 tokenizer = AutoTokenizer.from_pretrained(MODEL_PATH)
 model = AutoModelForCausalLM.from_pretrained(MODEL_PATH, device_map="auto")
@@ -101,6 +101,39 @@ out = model.generate(**generate_kwargs)
 print(tokenizer.decode(out[0][input_len:], skip_special_tokens=True))
 ```
 
+### vLLM Lib(0.6.4 and later version) for inference:
+
+```Python
+from transformers import AutoTokenizer
+from vllm import LLM, SamplingParams
+
+# THUDM/glm-4-9b-chat-1m-hf
+# max_model_len, tp_size = 1048576, 4
+# If you encounter OOM phenomenon, it is recommended to reduce max_model_len or increase tp_size
+max_model_len, tp_size = 131072, 1
+model_name = "THUDM/glm-4-9b-chat-1m-hf"
+prompt = [{"role": "user", "content": "what is your name?"}]
+
+tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True)
+llm = LLM(
+    model=model_name,
+    tensor_parallel_size=tp_size,
+    max_model_len=max_model_len,
+    trust_remote_code=True,
+    enforce_eager=True,
+    # GLM-4-9B-Chat-1M-HF If you encounter OOM phenomenon, it is recommended to enable the following parameters
+    # enable_chunked_prefill=True,
+    # max_num_batched_tokens=8192
+)
+stop_token_ids = [151329, 151336, 151338]
+sampling_params = SamplingParams(temperature=0.95, max_tokens=1024, stop_token_ids=stop_token_ids)
+
+inputs = tokenizer.apply_chat_template(prompt, tokenize=False, add_generation_prompt=True)
+outputs = llm.generate(prompts=inputs, sampling_params=sampling_params)
+
+print(outputs[0].outputs[0].text)
+```
+
 ## LICENSE
 
 The weights of the GLM-4 model are available under the terms of [LICENSE](LICENSE)