Skip to main content
This page is a work in progress. Rates shown here are not official. For current pricing, see telnyx.com/pricing/inference-api.
Pay-per-token. No minimums, no commitments.
CategoryBasisNotes
Text generationPer 1M tokens (input + output)Input and output priced separately
Audio transcriptionPer second of audioVaries by model
Text-to-speechPer 1M charactersVaries by voice/model
EmbeddingsPer 1M tokensSingle rate

​
Cached Tokens

Prompt caching on supported models. Cached input tokens at a discount.