model-api-hub 0.1.0

Unified Python interface for multiple AI model APIs

Model API Hub

One Line of Code, Access to 100+ AI Models

English · 简体中文

---

What is Model API Hub?

Model API Hub is a unified Python SDK that lets you access multiple AI model APIs across different platforms and modalities with a consistent, simple interface.

Stop juggling different SDKs for each provider. Use one library for everything.

# Same interface, different providers
from model_api_hub import deepseek_chat, siliconflow_chat, kimi_chat

# All work the same way
response = deepseek_chat("Hello!")
response = siliconflow_chat("Hello!")
response = kimi_chat("Hello!")

---

Key Features

Feature	Description
18+ LLM Providers	OpenAI, Anthropic, DeepSeek, ZhipuAI, Kimi, SiliconFlow, and more
Streaming Support	Real-time streaming responses for all major providers
5 Modalities	LLM, Vision-Language, Image Gen, Audio TTS, Video Gen
One-Line Setup	pip install model-api-hub and you're ready
Unified API	Same interface across all providers
Flexible Config	.env, YAML, or direct API keys
CLI Included	Test models directly from command line
Type Hints	Full type safety support

---

Quick Start

Installation

pip install model-api-hub

1. Set Your API Key

# Create .env file
echo 'DEEPSEEK_API_KEY=your_key_here' > .env

2. Start Coding

from model_api_hub import deepseek_chat

# That's it. You're done.
response = deepseek_chat("Explain quantum computing in simple terms")
print(response)

---

Usage Examples

Language Models (LLM)

Synchronous Chat

from model_api_hub import deepseek_chat, kimi_chat, siliconflow_chat

# DeepSeek
response = deepseek_chat(
    "Write a Python function to sort a list",
    system_prompt="You are a coding expert."
)

# Kimi (Moonshot)
response = kimi_chat(
    "Summarize this article",
    temperature=0.5
)

# SiliconFlow - access 50+ models
response = siliconflow_chat("Hello!", model="deepseek-ai/DeepSeek-V3")

Streaming Chat

from model_api_hub import deepseek_chat_stream

# Stream responses in real-time
for chunk in deepseek_chat_stream("Tell me a long story"):
    print(chunk, end="", flush=True)

Multi-turn Conversation

from model_api_hub.api.llm.deepseek_llm import create_client, get_completion

client = create_client()
messages = [
    {"role": "system", "content": "You are a helpful assistant."},
    {"role": "user", "content": "What is Python?"},
    {"role": "assistant", "content": "Python is a programming language..."},
    {"role": "user", "content": "What are its main features?"}
]

response = get_completion(client, messages)

Vision-Language Models (VLM)

from model_api_hub.api.vlm.openai_vlm import chat

response = chat(
    prompt="What's in this image?",
    image_path="photo.jpg"
)

Image Generation

from model_api_hub.api.image.siliconflow_image_gen import generate

image_url = generate("A beautiful sunset over mountains")

Text-to-Speech

from model_api_hub.api.audio.openai_tts import synthesize

audio = synthesize("Hello, world!", voice="alloy", output_path="hello.mp3")

---

Supported Models

Language Models (LLM)

• DeepSeek-Chat • DeepSeek-R1 • DeepSeek-Coder-V2 • DeepSeek-V3 • GPT-4o • GPT-4o-mini • GPT-4-Turbo • Claude-Sonnet-4.5 • Claude-Opus-4 • Gemini-Pro • Gemini-Flash • GLM-4.7-Flash • GLM-4 • GLM-4-Plus • GLM-4.5-Air • GLM-4.1-Thinking • Moonshot-v1-128k • Moonshot-v1-32k

• Kimi-K2 • MiniMax-ABAB6.5s • MiniMax-M2 • ERNIE-4.0 • ERNIE-4.5 • Qwen-Max • Qwen-Plus • Qwen-Turbo • Qwen2.5 • Qwen2.5-Coder • Qwen3 • Qwen2 • Qwen 1.5 • Llama4 • Llama3.1 • Llama3-70B • Gemma3 • Gemma-2

• Mistral-Large • Mixtral-8x22B • Command-R-Plus • InternLM3 • InternLM • InternLM2-20B • Baichuan • Yi 零一万物 • Yuan2.0 • Yuan2.0-M32 • Hunyuan-A13B • Hunyuan3D-2 • Spark-v3.5 • Phi4 • Phi-3 • MiniCPM • CharacterGLM

Vision-Language Models (VLM)

• GPT-4o-Vision • GPT-4V • Gemini-Pro-Vision • Qwen3-VL • Qwen2-VL • Qwen-VL-Plus

• GLM-4V • MiniCPM-o • Yi-VL • InternVL • DeepSeek-VL

• SpatialLM • LLaVA • CogVLM • BlueLM-Vision

Image Generation Models

• DALL-E 3 • DALL-E 2 • Kolors • Stable Diffusion XL • Stable Diffusion 3

• Recraft-v3 • Wanx • ERNIE-ViLG • Jimeng (Dreamina) • CogView

• Hunyuan-Image • Playground-v2 • Kandinsky • DeepFloyd IF

Audio Models

• Whisper • Whisper-Large-v3 • TTS-1 • TTS-1-HD • ElevenLabs-Multilingual-v2

• ElevenLabs-Flash • Azure-TTS • Azure-Speech • MiniMax-TTS • Baidu-TTS

• Qwen-Audio • ChatTTS • Fish-Speech • GPT-SoVITS

Video Generation Models

• Runway-Gen3 • Runway-Gen2 • Luma-Dream-Machine • Luma-Genie

• Pika • Stable-Video-Diffusion • Jimeng-Video • CogVideo

• VideoCrafter • ModelScope-Video • AnimateDiff

---

Configuration

Environment Variables (.env)

Create a .env file in your project root:

# LLM Providers
OPENAI_API_KEY=sk-...
ANTHROPIC_API_KEY=sk-...
DEEPSEEK_API_KEY=sk-...
KIMI_API_KEY=sk-...
ZHIPUAI_API_KEY=...
SILICONFLOW_API_KEY=sk-...
MINIMAX_API_KEY=...
YIYAN_API_KEY=...
DASHSCOPE_API_KEY=sk-...
MODELSCOPE_API_KEY=ms-...
XUNFEI_SPARK_API_KEY=...
GROQ_API_KEY=gsk_...
TOGETHER_API_KEY=...
MISTRAL_API_KEY=...
COHERE_API_KEY=...
PERPLEXITY_API_KEY=pplx-...
AZURE_OPENAI_API_KEY=...

# Other Services
ELEVENLABS_API_KEY=...
AZURE_SPEECH_KEY=...
STABILITY_API_KEY=...
RECRAFT_API_KEY=...
RUNWAY_API_KEY=...
LUMA_API_KEY=...

YAML Configuration

Create config.yaml:

llm:
  openai:
    model: "gpt-4o"
    temperature: 0.7
    max_tokens: 4096
  
  deepseek:
    model: "deepseek-chat"
    temperature: 0.7
    max_tokens: 4096

vlm:
  openai:
    model: "gpt-4o"
    
image:
  siliconflow:
    model: "Kwai-Kolors/Kolors"
    size: "1024x1024"

---

Documentation

• LLM Usage Guide - Complete LLM documentation
• API Reference - Full API reference
• llm.txt - Quick reference for AI assistants

---

Testing

Run tests for all providers:

# Test all LLMs (sync)
python tests/test_llm.py

# Test streaming
python tests/test_llm_streaming.py

# Test other modalities
python tests/test_vlm.py
python tests/test_image.py
python tests/test_audio.py
python tests/test_video.py

---

CLI Usage

# Chat with a provider
model-api-hub chat deepseek "Hello!"

# List available providers
model-api-hub list

# Test a provider
model-api-hub test deepseek

---

Architecture

model_api_hub/
├── api/
│   ├── llm/           # Language Models (18+ providers)
│   ├── vlm/           # Vision-Language Models
│   ├── image/         # Image Generation
│   ├── audio/         # Text-to-Speech
│   └── video/         # Video Generation
├── utils/
│   └── config.py      # Configuration management
├── cli.py             # Command-line interface
└── __init__.py        # Public API exports

---

Contributing

We welcome contributions! See CONTRIBUTING.md for guidelines.

Adding a New Provider

1. Create a new file in model_api_hub/api/llm/{provider}_llm.py
2. Implement chat(), chat_stream() (optional), and create_client()
3. Add exports to model_api_hub/api/llm/__init__.py
4. Add tests in tests/test_llm.py
5. Update documentation

See llm.txt for detailed implementation guide.

---

License

MIT License - see LICENSE file.

---

Support

• 📖 Documentation
• 🐛 Issue Tracker
• 💬 Discussions

---

Acknowledgments

Thanks to all the AI providers for their amazing APIs!