74 lines
2.3 KiB
Python
74 lines
2.3 KiB
Python
import gradio as gr
|
|
import requests
|
|
|
|
API_URL = "http://localhost:30000/generate" # ✅ 使用原生 generate 接口
|
|
API_KEY = "token-abc123"
|
|
MODEL_NAME = "Qwen3-14b-base"
|
|
|
|
# 🚫 不再拼接上下文，只保留用户当前输入
|
|
def build_prompt(history, user_message):
|
|
return user_message
|
|
|
|
# 主对话函数
|
|
def chat(user_message, history, max_tokens, temperature):
|
|
prompt = build_prompt(history, user_message)
|
|
|
|
headers = {
|
|
"Authorization": f"Bearer {API_KEY}",
|
|
"Content-Type": "application/json"
|
|
}
|
|
payload = {
|
|
"model": MODEL_NAME,
|
|
"prompt": prompt,
|
|
"max_tokens": max_tokens,
|
|
"temperature": temperature
|
|
}
|
|
|
|
try:
|
|
response = requests.post(API_URL, headers=headers, json=payload, timeout=30)
|
|
result = response.json()
|
|
reply = result["text"].strip() # ✅ /generate 接口返回字段是 text
|
|
except Exception as e:
|
|
reply = f"[请求失败] {e}"
|
|
|
|
return reply
|
|
|
|
# 测试 API 连通性
|
|
def test_api_connection(max_tokens, temperature):
|
|
headers = {
|
|
"Authorization": f"Bearer {API_KEY}",
|
|
"Content-Type": "application/json"
|
|
}
|
|
payload = {
|
|
"model": MODEL_NAME,
|
|
"prompt": "Ping?",
|
|
"max_tokens": max_tokens,
|
|
"temperature": temperature
|
|
}
|
|
|
|
try:
|
|
resp = requests.post(API_URL, headers=headers, json=payload, timeout=10)
|
|
out = resp.json()["text"].strip() # ✅ 修改这里
|
|
return f"✅ API 可用，响应: {out}"
|
|
except Exception as e:
|
|
return f"❌ API 请求失败: {e}"
|
|
|
|
# Gradio 界面
|
|
with gr.Blocks(title="Base 模型测试 UI") as demo:
|
|
gr.Markdown("# 💬 Base 模型对话界面")
|
|
|
|
with gr.Row():
|
|
max_tokens = gr.Slider(32, 1024, value=256, label="max_tokens")
|
|
temperature = gr.Slider(0.0, 1.5, value=0.7, step=0.05, label="temperature")
|
|
test_btn = gr.Button("🔁 测试 API 可用性")
|
|
test_output = gr.Textbox(label="API 测试结果", interactive=False)
|
|
|
|
chatbot = gr.ChatInterface(
|
|
fn=lambda msg, hist: chat(msg, hist, max_tokens.value, temperature.value),
|
|
title=None
|
|
)
|
|
|
|
test_btn.click(fn=test_api_connection, inputs=[max_tokens, temperature], outputs=test_output)
|
|
|
|
demo.launch(server_name="0.0.0.0", server_port=30001)
|