File size: 6,611 Bytes
bd06cc8
 
e100b79
bd06cc8
 
e100b79
bd06cc8
e100b79
 
 
 
395ec56
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
bd06cc8
 
 
 
 
 
 
 
 
 
 
e100b79
bd06cc8
 
 
 
 
 
 
 
 
e100b79
bd06cc8
fed1648
364bda2
 
c2b4e5c
364bda2
 
c2b4e5c
364bda2
 
fed1648
bd06cc8
 
 
 
bf3f697
bd06cc8
fed1648
bd06cc8
fed1648
 
 
 
 
bd06cc8
fed1648
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
b3ba3af
 
 
dd193a1
fed1648
2180807
dd193a1
fed1648
b3ba3af
 
 
 
 
fed1648
 
dd193a1
e100b79
b3ba3af
dd193a1
 
837e868
bd06cc8
 
e100b79
bd06cc8
fed1648
bd06cc8
 
 
 
 
 
 
fed1648
 
bd06cc8
 
 
 
 
 
 
 
 
 
 
 
fed1648
bd06cc8
 
 
 
 
e100b79
bd06cc8
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
import json
import sseclient
import requests
from flask import Flask, request, Response, stream_with_context
import random

app = Flask(__name__)

def generate_random_ip():
    return f"{random.randint(1,255)}.{random.randint(0,255)}.{random.randint(0,255)}.{random.randint(0,255)}"

def generate_user_agent():
    os_list = ['Windows NT 10.0', 'Windows NT 6.1', 'Mac OS X 10_15_7', 'Ubuntu', 'Linux x86_64']
    browser_list = ['Chrome', 'Firefox', 'Safari', 'Edge']
    chrome_version = f"{random.randint(70, 126)}.0.{random.randint(1000, 9999)}.{random.randint(100, 999)}"
    firefox_version = f"{random.randint(70, 100)}.0"
    safari_version = f"{random.randint(600, 615)}.{random.randint(1, 9)}.{random.randint(1, 9)}"
    edge_version = f"{random.randint(80, 100)}.0.{random.randint(1000, 9999)}.{random.randint(100, 999)}"

    os = random.choice(os_list)
    browser = random.choice(browser_list)

    if browser == 'Chrome':
        return f"Mozilla/5.0 ({os}) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/{chrome_version} Safari/537.36"
    elif browser == 'Firefox':
        return f"Mozilla/5.0 ({os}; rv:{firefox_version}) Gecko/20100101 Firefox/{firefox_version}"
    elif browser == 'Safari':
        return f"Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/{safari_version} (KHTML, like Gecko) Version/{safari_version.split('.')[0]}.1.2 Safari/{safari_version}"
    elif browser == 'Edge':
        return f"Mozilla/5.0 ({os}) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/{edge_version} Safari/537.36 Edg/{edge_version}"

def format_openai_response(content, finish_reason=None):
    return {
        "id": "chatcmpl-123",
        "object": "chat.completion.chunk",
        "created": 1677652288,
        "model": "gpt-4o",
        "choices": [{
            "delta": {"content": content} if content else {"finish_reason": finish_reason},
            "index": 0,
            "finish_reason": finish_reason
        }]
    }

@app.route('/hf/v1/chat/completions', methods=['POST'])
def chat_completions():
    data = request.json
    messages = data.get('messages', [])
    stream = data.get('stream', False)
    
    if not messages:
        return {"error": "No messages provided"}, 400
    
    model = data.get('model', 'gpt-4o')

    if model.startswith('gpt'):
        endpoint = "openAI"
        original_api_url = 'https://chatpro.ai-pro.org/api/ask/openAI'
    elif model.startswith('claude'):
        endpoint = "claude"
        original_api_url = 'https://chatpro.ai-pro.org/api/ask/claude'
    else:
        return {"error": "Unsupported model"}, 400

    headers = {
        'content-type': 'application/json',
        'X-Forwarded-For': generate_random_ip(),
        'origin': 'https://chatpro.ai-pro.org',
        'user-agent': generate_user_agent()
    }

    def generate():
        nonlocal messages
        full_response = ""
        while True:
            conversation = "\n".join([f"{msg['role']}: {msg['content']}" for msg in messages])
            conversation += "\n请关注并回复user最近的消息并避免总结对话历史的回答"
            
            payload = {
                "text": conversation,
                "endpoint": endpoint,
                "model": model
            }
            
            response = requests.post(original_api_url, headers=headers, json=payload, stream=True)
            client = sseclient.SSEClient(response)
            
            for event in client.events():
                if event.data.startswith('{"text":'):
                    data = json.loads(event.data)
                    new_content = data['text'][len(full_response):]
                    full_response = data['text']
                    
                    if new_content:
                        yield f"data: {json.dumps(format_openai_response(new_content))}\n\n"
                
                elif '"final":true' in event.data:
                    final_data = json.loads(event.data)
                    response_message = final_data.get('responseMessage', {})
                    finish_reason = response_message.get('finish_reason', 'stop')
                    
                    if finish_reason == 'length':
                        messages.append({"role": "assistant", "content": full_response})
                        messages.append({"role": "user", "content": "请继续你的输出,不要重复之前的内容"})
                        break  # 跳出当前循环,继续下一次请求
                    else:
                        # 正常结束,发送最后的内容(如果有的话)
                        last_content = response_message.get('text', '')
                        if last_content and last_content != full_response:
                            yield f"data: {json.dumps(format_openai_response(last_content[len(full_response):]))}\n\n"
                        
                        yield f"data: {json.dumps(format_openai_response('', finish_reason))}\n\n"
                        yield "data: [DONE]\n\n"
                        return  # 完全结束生成

        # 如果因为多次长度限制而最终结束,发送一个停止信号
        yield f"data: {json.dumps(format_openai_response('', 'stop'))}\n\n"
        yield "data: [DONE]\n\n"

    if stream:
        return Response(stream_with_context(generate()), content_type='text/event-stream')
    else:
        full_response = ""
        finish_reason = "stop"
        for chunk in generate():
            if chunk.startswith("data: ") and not chunk.strip() == "data: [DONE]":
                response_data = json.loads(chunk[6:])
                if 'choices' in response_data and response_data['choices']:
                    delta = response_data['choices'][0].get('delta', {})
                    if 'content' in delta:
                        full_response += delta['content']
                    if 'finish_reason' in delta:
                        finish_reason = delta['finish_reason']

        return {
            "id": "chatcmpl-123",
            "object": "chat.completion",
            "created": 1677652288,
            "model": model,
            "choices": [{
                "index": 0,
                "message": {
                    "role": "assistant",
                    "content": full_response
                },
                "finish_reason": finish_reason
            }],
            "usage": {
                "prompt_tokens": 0,
                "completion_tokens": 0,
                "total_tokens": 0
            }
        }

if __name__ == '__main__':
    app.run(debug=True, port=5000)