Models & Pricing

Browse the models available through TokenFly and how they are billed.

Available models

TokenFly brings China's top models behind one OpenAI-compatible API. For the complete, always-current list with live per-token prices, see the Models page.

FamilyExample modelsType
DeepSeekdeepseek-v4-flash, deepseek-v4-proText / reasoning
GLMGLM-5, glm-5.1, glm-5.2Text / reasoning
MiniMaxMiniMax-M3Text / reasoning
Kimikimi-k2.6Text / reasoning

Model names are passed in the model field of your request. Switch models by changing that one string.

List models programmatically

Fetch the live list of available model ids with a standard OpenAI-style call:

curl https://tokenfly666.com/v1/models \
  -H "Authorization: Bearer $TOKENFLY_API_KEY"

How billing works

  • Pay-as-you-go. You top up a prepaid balance and are charged per use.
  • Text models are billed per token (input + output), at the rates shown on the Models page.
  • No subscription and no monthly minimums.

Top up a prepaid balance to start; you only pay for what you use.

Choosing a model

  • Cheapest capable chat: deepseek-v4-flash, kimi-k2.6
  • Strong reasoning: deepseek-v4-pro, glm-5.2, MiniMax-M3
  • Balanced: GLM-5, glm-5.1