Introduce LLM class for offline inference (#115)

examples/offline_inference.py

@@ -0,0 +1,23 @@
+from cacheflow import LLM, SamplingParams
+
+
+# Sample prompts.
+prompts = [
+    "Hello, my name is",
+    "The president of the United States is",
+    "The capital of France is",
+    "The future of AI is",
+]
+# Create a sampling params object.
+sampling_params = SamplingParams(temperature=0.8, top_p=0.95)
+
+# Create an LLM.
+llm = LLM(model="facebook/opt-125m")
+# Generate texts from the prompts. The output is a list of RequestOutput objects
+# that contain the prompt, generated text, and other information.
+outputs = llm.generate(prompts, sampling_params)
+# Print the outputs.
+for output in outputs:
+    prompt = output.prompt
+    generated_text = output.outputs[0].text
+    print(f"Prompt: {prompt!r}, Generated text: {generated_text!r}")

examples/simple_server.py

@@ -1,13 +1,13 @@
 import argparse
 import uuid
 
-from cacheflow import (add_server_arguments, initialize_server_from_args,
-                       SamplingParams)
+from cacheflow import ServerArgs, LLMServer, SamplingParams
 
 
 def main(args: argparse.Namespace):
-    # Initialize the server.
-    server = initialize_server_from_args(args)
+    # Parse the CLI argument and initialize the server.
+    server_args = ServerArgs.from_cli_args(args)
+    server = LLMServer.from_server_args(server_args)
 
     # Test the following prompts.
     test_prompts = [
@@ -39,6 +39,6 @@ def main(args: argparse.Namespace):
 
 if __name__ == '__main__':
     parser = argparse.ArgumentParser(description='Simple CacheFlow server.')
-    parser = add_server_arguments(parser)
+    parser = ServerArgs.add_cli_args(parser)
     args = parser.parse_args()
     main(args)