初始化项目，由ModelHub XC社区提供模型

Model: kaividlabs/qwen3-4b-awq
Source: Original Platform

README.md

@@ -0,0 +1,27 @@
+---
+license: mit
+language:
+- en
+---
+
+Inference
+
+```bash
+pip install vllm
+```
+
+```py
+import os
+from vllm import LLM, SamplingParams
+
+os.environ["LD_LIBRARY_PATH"] = "/usr/local/lib/python3.12/dist-packages/nvidia/cu13/lib:" + os.environ.get("LD_LIBRARY_PATH", "")
+os.environ["VLLM_WORKER_MULTIPROC_METHOD"] = "spawn"
+
+llm = LLM(model="kaividlabs/qwen3-4b-awq", quantization="awq_marlin")
+sampling_params = SamplingParams(temperature=0.7, top_p=0.8, max_tokens=512)
+
+messages = [{"role": "user", "content": "Give me a short introduction to Formula-1"}]
+prompt = llm.get_tokenizer().apply_chat_template(messages, tokenize=False, add_generation_prompt=True, enable_thinking=False)
+
+print(outputs[0].outputs[0].text)
+```