What is Token?
The billing and budgeting atom of language models: text is split into subword pieces and every input, cached, thinking, and output token is metered separately. BYOB counts tokens with tiktoken and converts usage into credits per model pricing.
Example
A BYOB chat turn records input, cached, thinking, and output tokens separately, then prices each leg from the active model config — the same counters that feed the token meter and credit billing.
What people get wrong
Treating one token as one word or one character. Subword splitting means code and non-English text cost more tokens per character than plain English prose.
Frequently asked questions
Why do identical prompts cost different credits on different models?
Each model config carries its own input, output, cached, and thinking prices, so the same tokens convert to different credit amounts per model.
Do cached tokens still cost anything?
Yes, at a reduced cached rate — reuse lowers the bill instead of removing it.
Related terms
Token Metering
Per-leg usage accounting that turns raw token counts into billable credits using each model price table plus markup. BYOB meters input, output, cached, and thinking tokens separately, with a minimum charge flooring tiny calls.
Context Threshold
The input-size line where a provider switches a model to higher long-context pricing. BYOB model configs store the threshold with above-threshold rates, so sessions crossing it pay more per token without changing models.
AI Credits
The metered unit for AI work in BYOB: planning, generation, and agent runs draw down a balance tracked per project. Credits make heavy context work visible instead of hiding it in a flat fee.
Billing Action
The structured paywall signal (insufficient credits, quota exceeded, plan required) with a checkout URL. Generated apps use the same pattern to gate their own premium features.
Temperature
Sampling randomness for generation: low values make output focused and repeatable, high values make it varied and surprising. BYOB chat configs default near 0.5 while tool calls use 0.3 for steadier structured actions.
Thinking Budget
The token allowance reserved for internal model reasoning before answering. Larger budgets suit hard multi-step problems; BYOB prices thinking tokens separately so runaway reasoning shows up in metering.