File size: 10,736 Bytes
3079197 484e5ab 3079197 4c52eb9 9bf75d4 4c52eb9 3079197 f666f56 3079197 4c52eb9 8887e47 4c52eb9 2edbd4b 4c52eb9 f666f56 3079197 4c52eb9 b47e49a 2edbd4b 8887e47 4c52eb9 8887e47 4c52eb9 3079197 c1bdfb8 1ed30a6 c1bdfb8 1ed30a6 c1bdfb8 3079197 5e0a689 3079197 4a858d3 3079197 4a858d3 3079197 4a858d3 3079197 4a858d3 3079197 4a858d3 3079197 4a858d3 3079197 4a858d3 3079197 5e0a689 3079197 4a858d3 3079197 4a858d3 3079197 9fe9fc4 3079197 4a858d3 3079197 4a858d3 3079197 a8294f2 484e5ab 3079197 5e0a689 a8294f2 5e0a689 a8294f2 5e0a689 a8294f2 5e0a689 a8294f2 5e0a689 8887e47 5e0a689 9fe9fc4 004756c 9fe9fc4 7d85666 9fe9fc4 7d85666 9fe9fc4 3079197 c1bdfb8 3079197 c1bdfb8 3079197 1ed30a6 004756c 1ed30a6 004756c 1ed30a6 b085dec 004756c 1ed30a6 3079197 c1bdfb8 4c52eb9 3079197 c1bdfb8 3079197 b085dec |
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 |
#
# Copyright 2024 The InfiniFlow Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
#
import time
import uuid
from api.db import LLMType, UserTenantRole
from api.db.db_models import init_database_tables as init_web_db
from api.db.services import UserService
from api.db.services.llm_service import LLMFactoriesService, LLMService, TenantLLMService, LLMBundle
from api.db.services.user_service import TenantService, UserTenantService
from api.settings import CHAT_MDL, EMBEDDING_MDL, ASR_MDL, IMAGE2TEXT_MDL, PARSERS, LLM_FACTORY, API_KEY
def init_superuser():
user_info = {
"id": uuid.uuid1().hex,
"password": "admin",
"nickname": "admin",
"is_superuser": True,
"email": "[email protected]",
"creator": "system",
"status": "1",
}
tenant = {
"id": user_info["id"],
"name": user_info["nickname"] + "‘s Kingdom",
"llm_id": CHAT_MDL,
"embd_id": EMBEDDING_MDL,
"asr_id": ASR_MDL,
"parser_ids": PARSERS,
"img2txt_id": IMAGE2TEXT_MDL
}
usr_tenant = {
"tenant_id": user_info["id"],
"user_id": user_info["id"],
"invited_by": user_info["id"],
"role": UserTenantRole.OWNER
}
tenant_llm = []
for llm in LLMService.query(fid=LLM_FACTORY):
tenant_llm.append(
{"tenant_id": user_info["id"], "llm_factory": LLM_FACTORY, "llm_name": llm.llm_name, "model_type": llm.model_type,
"api_key": API_KEY})
if not UserService.save(**user_info):
print("\033[93m【ERROR】\033[0mcan't init admin.")
return
TenantService.insert(**tenant)
UserTenantService.insert(**usr_tenant)
TenantLLMService.insert_many(tenant_llm)
print("【INFO】Super user initialized. \033[93memail: [email protected], password: admin\033[0m. Changing the password after logining is strongly recomanded.")
chat_mdl = LLMBundle(tenant["id"], LLMType.CHAT, tenant["llm_id"])
msg = chat_mdl.chat(system="", history=[{"role": "user", "content": "Hello!"}], gen_conf={})
if msg.find("ERROR: ") == 0:
print("\33[91m【ERROR】\33[0m: ", "'{}' dosen't work. {}".format(tenant["llm_id"], msg))
embd_mdl = LLMBundle(tenant["id"], LLMType.EMBEDDING, tenant["embd_id"])
v, c = embd_mdl.encode(["Hello!"])
if c == 0:
print("\33[91m【ERROR】\33[0m:", " '{}' dosen't work!".format(tenant["embd_id"]))
factory_infos = [{
"name": "OpenAI",
"logo": "",
"tags": "LLM,TEXT EMBEDDING,SPEECH2TEXT,MODERATION",
"status": "1",
},{
"name": "Tongyi-Qianwen",
"logo": "",
"tags": "LLM,TEXT EMBEDDING,SPEECH2TEXT,MODERATION",
"status": "1",
},{
"name": "ZHIPU-AI",
"logo": "",
"tags": "LLM,TEXT EMBEDDING,SPEECH2TEXT,MODERATION",
"status": "1",
},
{
"name": "Local",
"logo": "",
"tags": "LLM,TEXT EMBEDDING,SPEECH2TEXT,MODERATION",
"status": "1",
},{
"name": "Moonshot",
"logo": "",
"tags": "LLM,TEXT EMBEDDING",
"status": "1",
}
# {
# "name": "文心一言",
# "logo": "",
# "tags": "LLM,TEXT EMBEDDING,SPEECH2TEXT,MODERATION",
# "status": "1",
# },
]
def init_llm_factory():
llm_infos = [
# ---------------------- OpenAI ------------------------
{
"fid": factory_infos[0]["name"],
"llm_name": "gpt-3.5-turbo",
"tags": "LLM,CHAT,4K",
"max_tokens": 4096,
"model_type": LLMType.CHAT.value
},{
"fid": factory_infos[0]["name"],
"llm_name": "gpt-3.5-turbo-16k-0613",
"tags": "LLM,CHAT,16k",
"max_tokens": 16385,
"model_type": LLMType.CHAT.value
},{
"fid": factory_infos[0]["name"],
"llm_name": "text-embedding-ada-002",
"tags": "TEXT EMBEDDING,8K",
"max_tokens": 8191,
"model_type": LLMType.EMBEDDING.value
},{
"fid": factory_infos[0]["name"],
"llm_name": "whisper-1",
"tags": "SPEECH2TEXT",
"max_tokens": 25*1024*1024,
"model_type": LLMType.SPEECH2TEXT.value
},{
"fid": factory_infos[0]["name"],
"llm_name": "gpt-4",
"tags": "LLM,CHAT,8K",
"max_tokens": 8191,
"model_type": LLMType.CHAT.value
},{
"fid": factory_infos[0]["name"],
"llm_name": "gpt-4-32k",
"tags": "LLM,CHAT,32K",
"max_tokens": 32768,
"model_type": LLMType.CHAT.value
},{
"fid": factory_infos[0]["name"],
"llm_name": "gpt-4-vision-preview",
"tags": "LLM,CHAT,IMAGE2TEXT",
"max_tokens": 765,
"model_type": LLMType.IMAGE2TEXT.value
},
# ----------------------- Qwen -----------------------
{
"fid": factory_infos[1]["name"],
"llm_name": "qwen-turbo",
"tags": "LLM,CHAT,8K",
"max_tokens": 8191,
"model_type": LLMType.CHAT.value
},{
"fid": factory_infos[1]["name"],
"llm_name": "qwen-plus",
"tags": "LLM,CHAT,32K",
"max_tokens": 32768,
"model_type": LLMType.CHAT.value
},{
"fid": factory_infos[1]["name"],
"llm_name": "qwen-max-1201",
"tags": "LLM,CHAT,6K",
"max_tokens": 5899,
"model_type": LLMType.CHAT.value
},{
"fid": factory_infos[1]["name"],
"llm_name": "text-embedding-v2",
"tags": "TEXT EMBEDDING,2K",
"max_tokens": 2048,
"model_type": LLMType.EMBEDDING.value
},{
"fid": factory_infos[1]["name"],
"llm_name": "paraformer-realtime-8k-v1",
"tags": "SPEECH2TEXT",
"max_tokens": 25*1024*1024,
"model_type": LLMType.SPEECH2TEXT.value
},{
"fid": factory_infos[1]["name"],
"llm_name": "qwen-vl-max",
"tags": "LLM,CHAT,IMAGE2TEXT",
"max_tokens": 765,
"model_type": LLMType.IMAGE2TEXT.value
},
# ---------------------- ZhipuAI ----------------------
{
"fid": factory_infos[2]["name"],
"llm_name": "glm-3-turbo",
"tags": "LLM,CHAT,",
"max_tokens": 128 * 1000,
"model_type": LLMType.CHAT.value
}, {
"fid": factory_infos[2]["name"],
"llm_name": "glm-4",
"tags": "LLM,CHAT,",
"max_tokens": 128 * 1000,
"model_type": LLMType.CHAT.value
}, {
"fid": factory_infos[2]["name"],
"llm_name": "glm-4v",
"tags": "LLM,CHAT,IMAGE2TEXT",
"max_tokens": 2000,
"model_type": LLMType.IMAGE2TEXT.value
},
{
"fid": factory_infos[2]["name"],
"llm_name": "embedding-2",
"tags": "TEXT EMBEDDING",
"max_tokens": 512,
"model_type": LLMType.EMBEDDING.value
},
# ---------------------- 本地 ----------------------
{
"fid": factory_infos[3]["name"],
"llm_name": "qwen-14B-chat",
"tags": "LLM,CHAT,",
"max_tokens": 4096,
"model_type": LLMType.CHAT.value
}, {
"fid": factory_infos[3]["name"],
"llm_name": "flag-embedding",
"tags": "TEXT EMBEDDING,",
"max_tokens": 128 * 1000,
"model_type": LLMType.EMBEDDING.value
},
# ------------------------ Moonshot -----------------------
{
"fid": factory_infos[4]["name"],
"llm_name": "moonshot-v1-8k",
"tags": "LLM,CHAT,",
"max_tokens": 7900,
"model_type": LLMType.CHAT.value
}, {
"fid": factory_infos[4]["name"],
"llm_name": "flag-embedding",
"tags": "TEXT EMBEDDING,",
"max_tokens": 128 * 1000,
"model_type": LLMType.EMBEDDING.value
},{
"fid": factory_infos[4]["name"],
"llm_name": "moonshot-v1-32k",
"tags": "LLM,CHAT,",
"max_tokens": 32768,
"model_type": LLMType.CHAT.value
},{
"fid": factory_infos[4]["name"],
"llm_name": "moonshot-v1-128k",
"tags": "LLM,CHAT",
"max_tokens": 128 * 1000,
"model_type": LLMType.CHAT.value
},
]
for info in factory_infos:
try:
LLMFactoriesService.save(**info)
except Exception as e:
pass
for info in llm_infos:
try:
LLMService.save(**info)
except Exception as e:
pass
"""
modify service_config
drop table llm;
drop table llm_factories;
update tenant_llm set llm_factory='Tongyi-Qianwen' where llm_factory='通义千问';
update tenant_llm set llm_factory='ZHIPU-AI' where llm_factory='智谱AI';
update tenant set parser_ids='naive:General,qa:Q&A,resume:Resume,manual:Manual,table:Table,paper:Paper,book:Book,laws:Laws,presentation:Presentation,picture:Picture,one:One';
alter table knowledgebase modify avatar longtext;
alter table user modify avatar longtext;
alter table dialog modify icon longtext;
"""
def init_web_data():
start_time = time.time()
if LLMFactoriesService.get_all().count() != len(factory_infos):
init_llm_factory()
if not UserService.get_all().count():
init_superuser()
print("init web data success:{}".format(time.time() - start_time))
if __name__ == '__main__':
init_web_db()
init_web_data() |