summaryrefslogtreecommitdiffstats
path: root/g4f/api/__init__.py
blob: d86364d182454635f80a37c3afffb25bfb2a9bb3 (plain) (blame)
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
157
158
159
160
161
162
163
164
165
166
167
from fastapi                  import FastAPI, Response, Request
from fastapi.middleware.cors  import CORSMiddleware
from typing                   import List, Union, Any, Dict, AnyStr
from ._tokenizer              import tokenize
import g4f
import time
import json
import random
import string
import uvicorn
import nest_asyncio

app = FastAPI()
nest_asyncio.apply()

origins = [
    "http://localhost",
    "http://localhost:1337",
]

app.add_middleware(
    CORSMiddleware,
    allow_origins=origins,
    allow_credentials=True,
    allow_methods=["*"],
    allow_headers=["*"],
)

JSONObject = Dict[AnyStr, Any]
JSONArray = List[Any]
JSONStructure = Union[JSONArray, JSONObject]

@app.get("/")
async def read_root():
    return Response(content=json.dumps({"info": "G4F API"}, indent=4), media_type="application/json")

@app.get("/v1")
async def read_root_v1():
    return Response(content=json.dumps({"info": "Go to /v1/chat/completions or /v1/models."}, indent=4), media_type="application/json")

@app.get("/v1/models")
async def models():
    model_list = [{
        'id': model,
        'object': 'model',
        'created': 0,
        'owned_by': 'g4f'} for model in g4f.Model.__all__()]

    return Response(content=json.dumps({
        'object': 'list',
        'data': model_list}, indent=4), media_type="application/json")

@app.get("/v1/models/{model_name}")
async def model_info(model_name: str):
    try:
        model_info = (g4f.ModelUtils.convert[model_name])

        return Response(content=json.dumps({
            'id': model_name,
            'object': 'model',
            'created': 0,
            'owned_by': model_info.base_provider
        }, indent=4), media_type="application/json")
    except:
        return Response(content=json.dumps({"error": "The model does not exist."}, indent=4), media_type="application/json")

@app.post("/v1/chat/completions")
async def chat_completions(request: Request, item: JSONStructure = None):

    item_data = {
        'model': 'gpt-3.5-turbo',
        'stream': False,
    }
    
    item_data.update(item or {})
    model = item_data.get('model')
    stream = item_data.get('stream')
    messages = item_data.get('messages')

    try:
        response = g4f.ChatCompletion.create(model=model, stream=stream, messages=messages)
    except:
        return Response(content=json.dumps({"error": "An error occurred while generating the response."}, indent=4), media_type="application/json")

    completion_id = ''.join(random.choices(string.ascii_letters + string.digits, k=28))
    completion_timestamp = int(time.time())

    if not stream:
        prompt_tokens, _ = tokenize(''.join([message['content'] for message in messages]))
        completion_tokens, _ = tokenize(response)

        json_data = {
            'id': f'chatcmpl-{completion_id}',
            'object': 'chat.completion',
            'created': completion_timestamp,
            'model': model,
            'choices': [
                {
                    'index': 0,
                    'message': {
                        'role': 'assistant',
                        'content': response,
                    },
                    'finish_reason': 'stop',
                }
            ],
            'usage': {
                'prompt_tokens': prompt_tokens,
                'completion_tokens': completion_tokens,
                'total_tokens': prompt_tokens + completion_tokens,
            },
        }

        return Response(content=json.dumps(json_data, indent=4), media_type="application/json")

    def streaming():
        try:
            for chunk in response:
                completion_data = {
                    'id': f'chatcmpl-{completion_id}',
                    'object': 'chat.completion.chunk',
                    'created': completion_timestamp,
                    'model': model,
                    'choices': [
                        {
                            'index': 0,
                            'delta': {
                                'content': chunk,
                            },
                            'finish_reason': None,
                        }
                    ],
                }

                content = json.dumps(completion_data, separators=(',', ':'))
                yield f'data: {content}\n\n'
                time.sleep(0.03)

            end_completion_data = {
                'id': f'chatcmpl-{completion_id}',
                'object': 'chat.completion.chunk',
                'created': completion_timestamp,
                'model': model,
                'choices': [
                    {
                        'index': 0,
                        'delta': {},
                        'finish_reason': 'stop',
                    }
                ],
            }

            content = json.dumps(end_completion_data, separators=(',', ':'))
            yield f'data: {content}\n\n'

        except GeneratorExit:
            pass

    return Response(content=json.dumps(streaming(), indent=4), media_type="application/json")

@app.post("/v1/completions")
async def completions():
    return Response(content=json.dumps({'info': 'Not working yet.'}, indent=4), media_type="application/json")

def run(ip):
    split_ip = ip.split(":")
    uvicorn.run(app, host=split_ip[0], port=int(split_ip[1]), use_colors=False)