fastmindapi 0.0.4

An easy-to-use, high-performance(?) backend for serving LLMs and other AI models, built on FastAPI.

FastMindAPI

An easy-to-use, high-performance(?) backend for serving LLMs and other AI models, built on FastAPI.

🚀 Quick Start

Install

pip install fastmindapi

Use

Run the server

# in Shell
fastmindapi-server --port 8000

# in Python
import fastmindapi as FM

server = FM.Server()
server.run()

Access via client / HTTP requests

curl http://IP:PORT/docs#/

import fastmindapi as FM

client = FM.Client(IP="x.x.x.x", PORT=xxx) # 127.0.0.1:8000 for default

client.add_model_info_list(model_info_list)
client.load_model(model_name)
client.generate(model_name, generation_request)

🪧 We primarily maintain the backend server; the client is provided for reference only. The main usage is through sending HTTP requests. (We might release FM-GUI in the future.)

✨ Features

Model: Support models with various backends

• ✅ Transformers

  • TransformersCausalLM ( AutoModelForCausalLM)
  • PeftCausalLM ( PeftModelForCausalLM )

• ✅ llama.cpp

  • LlamacppLM (Llama)

• MLC LLM

• vllm

• ...

Modules: More than just chatting with models

• Function Calling (extra tools in Python)
• Retrieval
• Agent
• ...

Flexibility: Easy to Use & Highly Customizable

• Load the model when coding / runtime
• Add any APIs you want