feat: add API streaming metrics and enhance tooltip UI with better loading animations
这个提交包含在:
@ -117,12 +117,17 @@ async def chat_endpoint(req: ChatRequest):
|
|||||||
|
|
||||||
client = AsyncOpenAI(api_key=api_key, base_url=base_url, timeout=10.0)
|
client = AsyncOpenAI(api_key=api_key, base_url=base_url, timeout=10.0)
|
||||||
|
|
||||||
|
# 调试日志:打印请求摘要
|
||||||
|
print(f"[API Request] Teacher: {req.teacher_type}, Stream: {req.stream}, Messages: {len(req.messages)}")
|
||||||
|
|
||||||
messages = [{"role": "system", "content": system_msg}]
|
messages = [{"role": "system", "content": system_msg}]
|
||||||
for msg in req.messages:
|
for msg in req.messages:
|
||||||
messages.append({"role": msg.role, "content": msg.content})
|
messages.append({"role": msg.role, "content": msg.content})
|
||||||
|
|
||||||
if req.stream:
|
if req.stream:
|
||||||
async def event_generator():
|
async def event_generator():
|
||||||
|
token_count = 0
|
||||||
|
start_time = asyncio.get_event_loop().time()
|
||||||
try:
|
try:
|
||||||
stream = await client.chat.completions.create(
|
stream = await client.chat.completions.create(
|
||||||
model=model_name,
|
model=model_name,
|
||||||
@ -132,6 +137,7 @@ async def chat_endpoint(req: ChatRequest):
|
|||||||
async for chunk in stream:
|
async for chunk in stream:
|
||||||
content = chunk.choices[0].delta.content
|
content = chunk.choices[0].delta.content
|
||||||
if content:
|
if content:
|
||||||
|
token_count += 1
|
||||||
# 转换成兼容 OpenAI 格式的 JSON,让前端解析逻辑更稳定
|
# 转换成兼容 OpenAI 格式的 JSON,让前端解析逻辑更稳定
|
||||||
data = {
|
data = {
|
||||||
"choices": [{
|
"choices": [{
|
||||||
@ -139,13 +145,18 @@ async def chat_endpoint(req: ChatRequest):
|
|||||||
}]
|
}]
|
||||||
}
|
}
|
||||||
yield f"data: {json.dumps(data, ensure_ascii=False)}\n\n"
|
yield f"data: {json.dumps(data, ensure_ascii=False)}\n\n"
|
||||||
|
|
||||||
|
duration = asyncio.get_event_loop().time() - start_time
|
||||||
|
print(f"[API Stream Success] Tokens: {token_count}, Time: {duration:.2f}s")
|
||||||
yield "data: [DONE]\n\n"
|
yield "data: [DONE]\n\n"
|
||||||
except Exception as e:
|
except Exception as e:
|
||||||
|
print(f"[API Stream Error] {str(e)}")
|
||||||
yield f"data: {json.dumps({'error': str(e)})}\n\n"
|
yield f"data: {json.dumps({'error': str(e)})}\n\n"
|
||||||
|
|
||||||
return StreamingResponse(event_generator(), media_type="text/event-stream")
|
return StreamingResponse(event_generator(), media_type="text/event-stream")
|
||||||
else:
|
else:
|
||||||
try:
|
try:
|
||||||
|
start_time = asyncio.get_event_loop().time()
|
||||||
response = await client.chat.completions.create(
|
response = await client.chat.completions.create(
|
||||||
model=model_name,
|
model=model_name,
|
||||||
messages=messages,
|
messages=messages,
|
||||||
|
|||||||
在新工单中引用
屏蔽一个用户