Overview

Python bindings for the LLM Providers registry. Provides zero-latency access to a curated dataset of LLM providers, models, pricing, and capabilities — powered by Rust via PyO3.

The registry is embedded in the native extension at build time from data/providers.json via build.rs (no runtime JSON parsing).

Features

• 🚀 Zero-Latency: Data is compiled into the binary; no runtime I/O or API calls.
• 🐍 Pythonic API: Simple functions and typed objects.
• 🔄 Unified Schema: Consistent data structure across all providers (OpenAI, Anthropic, DeepSeek, etc.).
• 📦 Rich Metadata: Includes pricing, context length, and tool support flags.

Installation

pip install llm-providers-list

Usage

import llm_providers_list

# 1. List all provider families
print(llm_providers_list.list_providers())
# Output: ['aliyun', 'anthropic', 'deepseek', 'openai', ...]

# 2. List all endpoint IDs (for direct configuration)
print(llm_providers_list.list_endpoints())
# Output: ['aliyun:cn', 'anthropic:global', 'moonshot:cn', 'moonshot:global', ...]

# 3. Get endpoint details by ID
family_id, ep = llm_providers_list.get_endpoint("moonshot:global")
print(f"Family: {family_id}, Base URL: {ep.base_url}, Region: {ep.region}")

# 4. Get specific model details
model = llm_providers_list.get_model("openai", "gpt-4o")
print(f"Model: {model.name}, Price: ${model.input_price}/1M tokens")

Supported Providers

• OpenAI (GPT-4o, GPT-3.5, o1)
• Anthropic (Claude 3.5 Sonnet, Haiku, Opus)
• DeepSeek (Chat, Reasoner)
• Aliyun (Qwen Max, Plus, Turbo)
• Tencent (Hunyuan)
• Moonshot (Kimi)
  • Moonshot AI (CN)
  • Moonshot AI Global
• MiniMax
  • MiniMax (CN)
  • MiniMax Global
• Zhipu (GLM-4)
  • BigModel (Zhipu CN)
  • Z.ai (Zhipu Global)
• Volcengine (Doubao)
• LongCat

Contributing

Contributions are welcome! See the main repository for details.

License

MIT