cochat-api

project name

Integrate the interfaces of various APIs

author

SchrodingersFish

version

v1.0.0

Project directory structure

| --- big_model large model
| --- Model object
| --- Utils Toolkit
| --- .env environment variables configuration file
| --- .env.example configuration file for environment variables
| --- .dockerignore files ignored during docker build
| --- DockerFile DockerFile build file
| --- README Project Description
| --- README_CN project Chinese description
| --- requirements.txt Dependencies required for project operation

1. How the source code runs

1. Install dependencies, install in order:

'pip install frozen list - 1.3.0 - py3 - none - any.whl'

'pip install multidict - 6.0.2 - py3 - none - any.whl'

'pip install -i https://pypi.tuna.tsinghua.edu.cn/simple -r requirements.txt'

2. Run the program:

'Python app.py'

2. How Docker Runs (Build Yourself)

1. Build an image 'Docker build -t cochat-api.'
2. Run the container 'Docker run -d -p 8000:8000 cochat-api'
3. Configurable environment variables are as follows

#iFLYTEK Spark configuration
-E SPARK_APPID = XXXX
-E SPARK_API_KEY = XXXX
-E SPARK_API_SECRET = XXXX
#Tongyi Qianwen configuration
-E DASHSCOPE_API_KEY = XXXX
#Wall Luca (front-end acquisition)
-E LUCA_ACCESS_TOKEN = XXXX
-E LUCA_CONVERSATION_ID = XXXX
#DeepSeek
-E DEEP_SEEK_API_KEY = XXXX
#MINIMAX
-E MINIMAX_API_KEY = XXXX
-E MINIMAX_GROUP_ID = XXXX
#LinkAI
-E LINK_AI_API_KEY = XXXX
#360 Brain
-E ZHINAO_API_KEY = XXXX
#chatanywhere
-E CHATANYWHERE_API_KEY = XXXX
#gemini
-E GEMINI_API_KEY=XXXX
#baichuan
-E BAICHUAN_API_KEY=XXXX

3. How Docker runs (using existing images on dockerhub)

1. Pull the mirror image 'Docker pull johnsonschrodinger/cochat-api: latest'
2. Configure environment variables, otherwise it cannot be used Refer to the known environment variables above
3. Run the container 'Docker run -d -p 8888:8000 johnsonschrodinger/cochat-api: latest'

4. How to use it

Integrate various common chat-APIs.

1. First call ip: port/api/login when using Get JWT_TOKEN "json {"Username": "admin", "password": "admin"} "..."

2. Then call according to JWT_TOKEN: ip: port/api/chat Header added: Authorization Bearer ${your JWT_TOKEN} "json {"Model": "qwen-max", "messages ": [ { " role": "user", "content": "What model are you " } ] } "..."

3. Current support models: 'Gpt-3.5-turbo ',' gpt-3.5-turbo-0125 ',' gpt-4 ',' deepseek-chat ',' deepseek-coder ',' gpt-3.5-turbo-16k ',' gpt-4 ',' wenxin ',' xunfei ',' luca ',' abab5.5-chat ',' abab5.5s-chat ',' abab6-chat ',' qwen-max ',' qwen-max-1201 ',' qwen-max-longcontext ',' spark ' (iFLYTEK Spark, model can be carried at the same level, version specified version),' 360gpt_s2_v9 ',' 360gpt-pro'

contact information

Email: schrodingersfish@outlook.com

Special instructions

The copyright of this project is protected by law, and dissemination is prohibited without the author's permission