Model Market
No model data
Frequently Asked Questions
How is model pricing calculated?
Priced per million tokens (¥/M tokens). Chat models have separate input and output prices, billed by actual token consumption. Some models are billed per request. Check the Model Market for real-time pricing.
How do I choose the right model?
Choose based on your needs: GPT-4o / Claude 3.5 Sonnet for complex reasoning and code generation; DeepSeek-V3 / Qwen-Max for cost-effective general conversation; GPT-4o-mini / DeepSeek-Chat for simple tasks at lower cost. Use the filters to narrow by type and capabilities.
What do the "Stream", "Functions", and "Vision" badges mean?
"Stream" means the model supports real-time token-by-token output (ideal for chat). "Functions" means it supports Tool/Function Calling (for AI Agent development). "Vision" means it can process and understand images.