These guides price LLM APIs the way a coding agent uses them: one whole session, not one token. Each one states the session it prices, works the cost out from published rates, and says what the figure leaves out. Last updated .
Every dollar figure on these pages is computed, not typed. Prices for Openference come from the public model catalog at each model's most expensive hour of the day, so a model with a scheduled discount costs less at other hours than shown. Prices for other gateways are the rates each one publishes, with the page they were read from and the day they were read.
The session is the same in every guide: forty turns, an opening prompt that grows each turn, and a fixed number of tokens written per turn. A guide that changes it says so. The figures assume no caching unless a column says otherwise, because cached rates differ by model and can reorder a table.
Where Openference is not the cheapest option for a model, the guide says that too. None of the rankings measure quality: a cheaper model that fails the task costs more than a dearer one that finishes it. Compare prices on the pricing page and the comparison index.
Model catalog · Pricing · Comparisons · Cost guides · Integration guides · FAQ
Guides: GLM coding plan alternative · Free AI API · LLM gateway · LLM proxy · Cheap AI API · OpenAI-compatible API
Model APIs: GLM API · DeepSeek API · Kimi API · Qwen API · MiniMax API