Glossary · Training & tuning

reasoning tokens

The intermediate "thinking" tokens a reasoning model generates before its final answer, often billed separately from output tokens.

Related terms

• Hugging Face Transformers
• LoRA (low-rank adaptation)
• QLoRA
• alignment (LLM)
• direct preference optimization (DPO)
• extended thinking

Vynaris is an inference gateway that routes each request to the cheapest right-sized model and shows the receipt. Get an API key or read the docs.