From bb0501c0d96cc9531dc8885dec8d5e701156010d Mon Sep 17 00:00:00 2001
From: yichuan~ <73766326+yichuan520030910320@users.noreply.github.com>
Date: Wed, 31 Jul 2024 04:40:51 +0800
Subject: [PATCH] Fix List input bug (#838)

---
 examples/usage/openai_batch_chat.py      | 10 ++++++++++
 examples/usage/openai_batch_complete.py  | 11 +++++++++++
 examples/usage/openai_parallel_sample.py | 20 ++++++++++++++++++--
 python/sglang/srt/openai_api/adapter.py  |  2 +-
 4 files changed, 40 insertions(+), 3 deletions(-)

diff --git a/examples/usage/openai_batch_chat.py b/examples/usage/openai_batch_chat.py
index cffa50c67..8640d0925 100644
--- a/examples/usage/openai_batch_chat.py
+++ b/examples/usage/openai_batch_chat.py
@@ -1,3 +1,13 @@
+"""
+Usage:
+python -m sglang.launch_server --model-path meta-llama/Llama-2-7b-chat-hf --port 30000
+python openai_batch_chat.py
+Note: Before running this script,
+you should create the input.jsonl file with the following content:
+{"custom_id": "request-1", "method": "POST", "url": "/v1/chat/completions", "body": {"model": "gpt-3.5-turbo-0125", "messages": [{"role": "system", "content": "You are a helpful assistant."},{"role": "user", "content": "Hello world!  List 3 NBA players and tell a story"}],"max_tokens": 300}}
+{"custom_id": "request-2", "method": "POST", "url": "/v1/chat/completions", "body": {"model": "gpt-3.5-turbo-0125", "messages": [{"role": "system", "content": "You are an assistant. "},{"role": "user", "content": "Hello world! List three capital and tell a story"}],"max_tokens": 500}}
+"""
+
 import json
 import os
 import time
diff --git a/examples/usage/openai_batch_complete.py b/examples/usage/openai_batch_complete.py
index 3cf2ede0b..af694b54a 100644
--- a/examples/usage/openai_batch_complete.py
+++ b/examples/usage/openai_batch_complete.py
@@ -1,3 +1,14 @@
+"""
+Usage:
+python -m sglang.launch_server --model-path meta-llama/Llama-2-7b-chat-hf --port 30000
+python openai_batch_complete.py
+Note: Before running this script,
+you should create the input.jsonl file with the following content:
+{"custom_id": "request-1", "method": "POST", "url": "/v1/completions", "body": {"model": "gpt-3.5-turbo-instruct", "prompt": "List 3 names of famous soccer player: ", "max_tokens": 200}}
+{"custom_id": "request-2", "method": "POST", "url": "/v1/completions", "body": {"model": "gpt-3.5-turbo-instruct", "prompt": "List 6 names of famous basketball player:  ", "max_tokens": 400}}
+{"custom_id": "request-3", "method": "POST", "url": "/v1/completions", "body": {"model": "gpt-3.5-turbo-instruct", "prompt": "List 6 names of famous basketball player:  ", "max_tokens": 400}}
+"""
+
 import json
 import os
 import time
diff --git a/examples/usage/openai_parallel_sample.py b/examples/usage/openai_parallel_sample.py
index 0d3a372b4..99ec0aa58 100644
--- a/examples/usage/openai_parallel_sample.py
+++ b/examples/usage/openai_parallel_sample.py
@@ -17,9 +17,9 @@ print(response)
 response = client.completions.create(
     model="default",
     prompt="I am a robot and I want to study like humans. Now let's tell a story. Once upon a time, there was a little",
-    n=1,
+    n=5,
     temperature=0.8,
-    max_tokens=32,
+    max_tokens=320,
 )
 print(response)
 
@@ -68,6 +68,22 @@ response = client.completions.create(
 print(response)
 
 
+response = client.completions.create(
+    model="default",
+    prompt=[
+        "prompt1: I am a robot and I want to learn like humans. Now let's begin a tale. Once upon a time, there was a small",
+        "prompt2: As a robot, my goal is to understand human learning. Let's start a story. In a faraway land, there lived a tiny",
+        "prompt3: Being a robot, I aspire to study like people. Let's share a story. Long ago, there was a little",
+        "prompt4: I am a robot aiming to learn like humans. Let's narrate a story. Once, in a distant kingdom, there was a young",
+        "prompt5: As a robot, I seek to learn in human ways. Let's tell a story. Once upon a time, in a small village, there was a young",
+    ],
+    n=1,
+    temperature=0.8,
+    max_tokens=320,
+)
+print(response)
+
+
 # Text completion
 response = client.completions.create(
     model="default",
diff --git a/python/sglang/srt/openai_api/adapter.py b/python/sglang/srt/openai_api/adapter.py
index 5fa75f1b8..21f38cd22 100644
--- a/python/sglang/srt/openai_api/adapter.py
+++ b/python/sglang/srt/openai_api/adapter.py
@@ -361,7 +361,7 @@ def v1_generate_request(all_requests):
     if len(all_requests) == 1:
         prompt = prompts[0]
         sampling_params_list = sampling_params_list[0]
-        if isinstance(prompts, str) or isinstance(prompts[0], str):
+        if isinstance(prompt, str) or isinstance(prompt[0], str):
             prompt_kwargs = {"text": prompt}
         else:
             prompt_kwargs = {"input_ids": prompt}