Introduction

Welcome to Tokensmart — a unified AI API gateway. With a single API key you can call nearly every major LLM available today, including:

• OpenAI: gpt-5.5, gpt-5.4, gpt-5.4-mini, gpt-image-2
• Anthropic: Claude Opus 4.8 / 4.7 · Sonnet 4.6 · Haiku 4.5
• Google: gemini-3-pro-preview, gemini-3.5-flash, gemini-2.5-pro
• DeepSeek: DeepSeek V4 (deepseek-v4-pro / deepseek-v4-flash)
• Chinese models: the Qwen family (qwen3.5-plus / qwen3-max / qwen3-vl-plus, …); Kimi, Zhipu and more coming soon

Key features

• Unified protocol: Whether you use the OpenAI SDK or the Anthropic SDK, you can call any model by just swapping base_url
• Pay-as-you-go: No monthly fees, no minimum commit. Top up from ¥1, pay only for what you use
• Transparent billing: Every request is logged individually with token count, latency, and cost
• Prompt caching: Anthropic prompt caching is fully supported — long system prompts cost 90% less
• Smart routing: Automatically fails over on errors for high availability

Next steps

• Quickstart — make your first API call in under 5 minutes
• Authentication — how to create and use API keys
• Chat Completions — reference for the chat endpoint