Model Catalog

All models are accessible via a single API key. Prices in ¥ (Chinese Yuan).

Text Generation

DeepSeek Chat

deepseek-chat
text

Fast, capable chat model for general-purpose text generation, summarization, and Q&A.

¥0.14 / ¥0.28 per 1M tokens
View details →

DeepSeek Chat (Search)

deepseek-chat-search
text

DeepSeek Chat with integrated web search. Answers include real-time information and citations.

¥0.14 / ¥0.28 per 1M tokens
View details →

DeepSeek Chat (Think)

deepseek-chat-think
text

DeepSeek Chat with deep thinking mode. Shows step-by-step reasoning process.

¥0.55 / ¥2.19 per 1M tokens
View details →

DeepSeek R1

deepseek-r1
text

Advanced reasoning model with Chain-of-Thought. Excels at math, coding, and complex analysis.

¥0.55 / ¥2.19 per 1M tokens
View details →

DeepSeek R1 (Search)

deepseek-r1-search
text

DeepSeek R1 with web search capabilities. Combines deep reasoning with real-time information.

¥0.55 / ¥2.19 per 1M tokens
View details →

Doubao (豆包)

doubao
text

ByteDance's Doubao model. Supports text chat and multimodal input including images and files.

¥0.10 / ¥0.20 per 1M tokens
View details →

Gemini Advanced

gemini-advanced
text

Google Gemini Advanced (Pro). Enhanced capabilities for complex tasks and analysis.

¥0.30 / ¥0.60 per 1M tokens
View details →

Gemini Web

gemini-web
text

Google Gemini base model accessed via web interface. Strong general reasoning.

¥0.15 / ¥0.30 per 1M tokens
View details →

Hailuo (海螺)

hailuo
text

Hailuo AI (MiniMax). Optimized for conversational Chinese with natural, fluent responses.

¥0.10 / ¥0.20 per 1M tokens
View details →

Kimi

kimi
text

Moonshot's Kimi model with built-in web search and long context support up to 128K tokens.

¥0.20 / ¥0.40 per 1M tokens
View details →

Kimi (Silent Search)

kimi-silent_search
text

Kimi with silent web search. Searches the web behind the scenes without showing search results.

¥0.20 / ¥0.40 per 1M tokens
View details →

MiniMax

minimax
text

MiniMax's flagship model with strong Chinese and English bilingual capabilities.

¥0.10 / ¥0.20 per 1M tokens
View details →

Qwen (通义千问)

qwen
text

Alibaba's Qwen (Tongyi Qianwen). Competitive performance across reasoning, coding, and multilingual tasks.

¥0.10 / ¥0.20 per 1M tokens
View details →

Ready to integrate?

All models share the same OpenAI-compatible API. See the docs to get started in minutes.