Update README.md

README.md

@@ -2,4 +2,41 @@
 license: apache-2.0
 ---
 This is a copy of Qwen3-8B compiled with the 2.25 SDK for the Neuron workshop.
-https://github.com/aws-neuron/neuron-workshops
+https://github.com/aws-neuron/neuron-workshops
+
+This checkpoint was generated with the code:
+```
+bs=1
+seqlength=1024
+
+import os
+from vllm import LLM, SamplingParams
+os.environ['VLLM_NEURON_FRAMEWORK'] = "neuronx-distributed-inference"
+
+path = f"/home/ubuntu/qwen3/qwen3-8B-BS{bs}-SEQ{seqlength}"
+#save the sharded weights and compiler artifacts in the same folder
+os.environ['NEURON_COMPILED_ARTIFACTS'] = path
+os.environ['BASE_COMPILE_WORK_DIR'] =path
+
+llm = LLM(
+    model="/home/ubuntu/models/Qwen3-8B",
+    max_num_seqs=bs,
+    max_model_len=seqlength,
+    device="neuron",
+    tensor_parallel_size=2,
+    override_neuron_config={"save_sharded_checkpoint": True})
+prompts = [
+    "Hello, my name is",
+    "The president of the United States is",
+    "The capital of France is",
+    "The future of AI is",
+]
+# note that top_k must be set to lower than the global_top_k defined in
+# the neuronx_distributed_inference.models.config.OnDeviceSamplingConfig
+sampling_params = SamplingParams(top_k=10, temperature=0.8, top_p=0.95)
+outputs = llm.generate(prompts, sampling_params)
+for output in outputs:
+    prompt = output.prompt
+    generated_text = output.outputs[0].text
+    print(f"Prompt: {prompt!r}, Generated text: {generated_text!r}")
+```