File size: 14,793 Bytes
c57ef37 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 |
import os
import time
import json
import random
import requests
import g4f
from fastapi import FastAPI,Response, status
from typing import Dict, NewType, Union, Optional, List, get_type_hints
from fastapi.middleware.cors import CORSMiddleware
from fastapi.responses import StreamingResponse,JSONResponse
from fastapi.openapi.utils import get_openapi
from starlette.exceptions import HTTPException as StarletteHTTPException
from pydantic import BaseModel
#app = FastAPI(docs_url=None, redoc_url=None)
app = FastAPI()
app.add_middleware(
CORSMiddleware,
allow_origins=["*"],
allow_methods=["*"],
allow_headers=["*"],
)
class chat_completions_Item(BaseModel):
stream : bool = False
model : str = 'gpt-3.5-turbo'
messages : list = [{'role': 'user', 'content':"Say 'Hello World'."}]
provider: Union[str, None] = None
temperature : float = 0.8
presence_penalty: float = 0
frequency_penalty: float = 0
top_p: float = 1
class completions_Item(BaseModel):
stream : bool = False
model : str = 'gpt-3.5-turbo'
prompt : str = "Say 'Hello World'."
provider : Union[str, None] = None
temperature : float = 0.8
presence_penalty: float = 0
frequency_penalty: float = 0
top_p: float = 1
def auto_select(model:str='gpt-3.5-turbo',stream:bool=False):
r = requests.get('https://gpt.lemonsoftware.eu.org/v1/status')
data = r.json()['data']
model_providers = set()
random.shuffle(data)
for provider_info in data:
for model_info in provider_info['model']:
if model in model_info:
model_providers.add(provider_info['provider'])
if model_info[model]['status'] == 'Active':
if stream == True and getattr(g4f.Provider,provider_info['provider']).supports_stream == False:
continue
return [getattr(g4f.Provider,provider_info['provider']),provider_info['provider']]
else:
continue
break
if not model_providers:
return None
active_providers = set()
for provider_info in data:
for model_info in provider_info['model']:
for model in model_info.values():
if model['status'] == 'Active':
if stream == True and getattr(g4f.Provider,provider_info['provider']).supports_stream == False:
continue
active_providers.add(provider_info['provider'])
chooseable_providers = model_providers & active_providers
if not chooseable_providers:
return None
chooseable_provider = random.choice(list(chooseable_providers))
return [getattr(g4f.Provider,chooseable_provider),chooseable_provider]
@app.post("/v1/chat/completions")
def chat_completions(item: chat_completions_Item,responses: Response):
stream = item.stream
model = item.model.lower()
messages = item.messages
provider_name = item.provider
temperature = item.temperature
presence_penalty = item.presence_penalty
frequency_penalty = item.frequency_penalty
top_p = item.top_p
if provider_name:
try:
response = g4f.ChatCompletion.create(model=model, provider=getattr(g4f.Provider,provider_name),stream=stream,messages=messages,temperature=temperature,presence_penalty=presence_penalty,frequency_penalty=frequency_penalty,top_p=top_p)
except:
return JSONResponse(status_code=status.HTTP_500_INTERNAL_SERVER_ERROR, content={"error": {"message": "There was an error","type": "invalid_request_error","param": None,"code": 500}})
else:
provider = auto_select(model=model,stream=stream)
if provider == None:
return JSONResponse(status_code=status.HTTP_500_INTERNAL_SERVER_ERROR, content={"error": {"message": "The model is invalid or not working.","type": "invalid_request_error","param": None,"code": 500}})
if stream and provider[0].supports_stream == False:
return JSONResponse(status_code=status.HTTP_500_INTERNAL_SERVER_ERROR, content={"error": {"message": "Stream is not supported.","type": "invalid_request_error","param": None,"code": 500}})
provider_name = provider[1]
try:
response = g4f.ChatCompletion.create(model=model, provider=provider[0],stream=stream,messages=messages,temperature=temperature,presence_penalty=presence_penalty,frequency_penalty=frequency_penalty,top_p=top_p)
except:
provider = auto_select(model=model,stream=stream)
if provider == None:
return JSONResponse(status_code=status.HTTP_500_INTERNAL_SERVER_ERROR, content={"error": {"message": "The model is invalid or not working.","type": "invalid_request_error","param": None,"code": 500}})
if stream and provider[0].supports_stream == False:
return JSONResponse(status_code=status.HTTP_500_INTERNAL_SERVER_ERROR, content={"error": {"message": "Stream is not supported.","type": "invalid_request_error","param": None,"code": 500}})
provider_name = provider[1]
try:
response = g4f.ChatCompletion.create(model=model, provider=provider[0],stream=stream,messages=messages,temperature=temperature,presence_penalty=presence_penalty,frequency_penalty=frequency_penalty,top_p=top_p)
except:
return JSONResponse(status_code=status.HTTP_500_INTERNAL_SERVER_ERROR, content={"error": {"message": "There was an error.Please try again.","type": "server_error","param": None,"code": 500}})
if not stream:
completion_timestamp = int(time.time())
completion_id = ''.join(random.choices(
'ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789', k=28))
return {
'id': 'chatcmpl-%s' % completion_id,
'object': 'chat.completion',
'created': completion_timestamp,
'model': model,
'provider':provider_name,
'usage': {
'prompt_tokens': len(messages),
'completion_tokens': len(response),
'total_tokens': len(messages)+len(response)
},
'choices': [{
'message': {
'role': 'assistant',
'content': response
},
'finish_reason': 'stop',
'index': 0
}]
}
def stream():
nonlocal response
for token in response:
completion_timestamp = int(time.time())
completion_id = ''.join(random.choices(
'ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789', k=28))
completion_data = {
'id': f'chatcmpl-{completion_id}',
'object': 'chat.completion.chunk',
'created': completion_timestamp,
'model': model,
'provider':provider_name,
'choices': [
{
'delta': {
'content': token
},
'index': 0,
'finish_reason': None
}
]
}
yield 'data: %s\n\n' % json.dumps(completion_data, separators=(',' ':'))
time.sleep(0.1)
return StreamingResponse(stream(), media_type='text/event-stream')
@app.post("/v1/completions")
def completions(item: completions_Item,responses: Response):
stream = item.stream
model = item.model.lower()
messages = [{'role': 'user', 'content':item.prompt}]
provider_name = item.provider
temperature = item.temperature
presence_penalty = item.presence_penalty
frequency_penalty = item.frequency_penalty
top_p = item.top_p
if provider_name:
try:
response = g4f.ChatCompletion.create(model=model, provider=getattr(g4f.Provider,provider_name),stream=stream,messages=messages,temperature=temperature,presence_penalty=presence_penalty,frequency_penalty=frequency_penalty,top_p=top_p)
except:
return JSONResponse(status_code=status.HTTP_500_INTERNAL_SERVER_ERROR, content={"error": {"message": "There was an error.","type": "invalid_request_error","param": None,"code": 500}})
else:
provider = auto_select(model=model,stream=stream)
if provider == None:
return JSONResponse(status_code=status.HTTP_500_INTERNAL_SERVER_ERROR, content={"error": {"message": "The model is invalid or not working.","type": "invalid_request_error","param": None,"code": 500}})
if stream and provider[0].supports_stream == False:
return JSONResponse(status_code=status.HTTP_500_INTERNAL_SERVER_ERROR, content={"error": {"message": "Stream is not supported.","type": "invalid_request_error","param": None,"code": 500}})
provider_name = provider[1]
try:
response = g4f.ChatCompletion.create(model=model, provider=provider[0],stream=stream,messages=messages,temperature=temperature,presence_penalty=presence_penalty,frequency_penalty=frequency_penalty,top_p=top_p)
except:
provider = auto_select(model=model,stream=stream)
if provider == None:
return JSONResponse(status_code=status.HTTP_500_INTERNAL_SERVER_ERROR, content={"error": {"message": "The model is invalid or not working.","type": "invalid_request_error","param": None,"code": 500}})
if stream and provider[0].supports_stream == False:
return JSONResponse(status_code=status.HTTP_500_INTERNAL_SERVER_ERROR, content={"error": {"message": "Stream is not supported.","type": "invalid_request_error","param": None,"code": 500}})
provider_name = provider[1]
try:
response = g4f.ChatCompletion.create(model=model, provider=provider[0],stream=stream,messages=messages,temperature=temperature,presence_penalty=presence_penalty,frequency_penalty=frequency_penalty,top_p=top_p)
except:
return JSONResponse(status_code=status.HTTP_500_INTERNAL_SERVER_ERROR, content={"error": {"message": "There was an error.Please try again.","type": "server_error","param": None,"code": 500}})
if not stream:
completion_timestamp = int(time.time())
completion_id = ''.join(random.choices(
'ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789', k=28))
return {
'id': 'cmpl-%s' % completion_id,
'object': 'text.completion',
'created': completion_timestamp,
'model': model,
'provider':provider_name,
'usage': {
'prompt_tokens': len(messages),
'completion_tokens': len(response),
'total_tokens': len(messages)+len(response)
},
'choices': [{
'text': response,
'finish_reason': 'length',
"logprobs": None,
'index': 0
}]
}
def stream():
nonlocal response
for token in response:
completion_timestamp = int(time.time())
completion_id = ''.join(random.choices(
'ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789', k=28))
completion_data = {
'id': f'cmpl-{completion_id}',
'object': 'text.completion.chunk',
'created': completion_timestamp,
'model': model,
'provider':provider_name,
'choices': [
{
'delta': {
'text': token
},
'index': 0,
'finish_reason': None
}
]
}
yield 'data: %s\n\n' % json.dumps(completion_data, separators=(',' ':'))
time.sleep(0.1)
return StreamingResponse(stream(), media_type='text/event-stream')
@app.get("/v1/dashboard/billing/subscription")
@app.get("/dashboard/billing/subscription")
async def billing_subscription():
return {
"object": "billing_subscription",
"has_payment_method": True,
"canceled": False,
"canceled_at": None,
"delinquent": None,
"access_until": 2556028800,
"soft_limit": 6944500,
"hard_limit": 166666666,
"system_hard_limit": 166666666,
"soft_limit_usd": 416.67,
"hard_limit_usd": 9999.99996,
"system_hard_limit_usd": 9999.99996,
"plan": {
"title": "Pay-as-you-go",
"id": "payg"
},
"primary": True,
"account_name": "OpenAI",
"po_number": None,
"billing_email": None,
"tax_ids": None,
"billing_address": {
"city": "New York",
"line1": "OpenAI",
"country": "US",
"postal_code": "NY10031"
},
"business_address": None
}
@app.get("/v1/dashboard/billing/usage")
@app.get("/dashboard/billing/usage")
async def billing_usage(start_date:str='2023-01-01',end_date:str='2023-01-31'):
return {
"object": "list",
"daily_costs": [
{
"timestamp": time.time(),
"line_items": [
{
"name": "GPT-4",
"cost": 0.0
},
{
"name": "Chat models",
"cost": 1.01
},
{
"name": "InstructGPT",
"cost": 0.0
},
{
"name": "Fine-tuning models",
"cost": 0.0
},
{
"name": "Embedding models",
"cost": 0.0
},
{
"name": "Image models",
"cost": 16.0
},
{
"name": "Audio models",
"cost": 0.0
}
]
}
],
"total_usage": 1.01
}
@app.get("/v1/models")
def models():
import g4f.models
model = {"data":[]}
for i in g4f.models.ModelUtils.convert:
model['data'].append({
"id": i,
"object": "model",
"owned_by": g4f.models.ModelUtils.convert[i].base_provider,
"tokens": 99999,
"fallbacks": None,
"endpoints": [
"/v1/chat/completions"
],
"limits": None,
"permission": []
})
return model
@app.get("/v1/providers")
async def providers():
files = os.listdir("g4f/Provider/Providers")
files = [f for f in files if os.path.isfile(os.path.join("g4f/Provider/Providers", f))]
files.sort(key=str.lower)
providers_data = {"data":[]}
for file in files:
if file.endswith(".py"):
name = file[:-3]
try:
p = getattr(g4f.Provider,name)
providers_data["data"].append({
"provider": str(name),
"model": list(p.model),
"url": str(p.url),
"working": bool(p.working),
"supports_stream": bool(p.supports_stream)
})
except:
pass
return providers_data
def custom_openapi():
if app.openapi_schema:
return app.openapi_schema
openapi_schema = get_openapi(
title="GPT API",
version="1.0.0",
summary="GPT API",
description="[Try Online](https://chatgpt-next-web.lemonsoftware.eu.org/)",
routes=app.routes,
)
openapi_schema["info"]["x-logo"] = {
"url": "https://gpt-status.lemonsoftware.eu.org/icon.svg"
}
app.openapi_schema = openapi_schema
return app.openapi_schema
app.openapi = custom_openapi
@app.exception_handler(StarletteHTTPException)
async def custom_http_exception_handler(request, exc):
return JSONResponse(status_code=status.HTTP_404_NOT_FOUND, content={"error": {"message": "Invalid URL","type": "invalid_request_error","param": None,"code": 404}})
if __name__ == '__main__':
import uvicorn
uvicorn.run(app, host="0.0.0.0", port=7860) |