Calculate combined API costs for text, image, and audio inputs in a multimodal request.
Multimodal AI requests often mix text, image, and audio inputs/outputs, each billed under a different pricing unit — text and images per token, audio per minute. This calculator sums the cost contribution of each modality independently to produce a single blended total, making it easier to budget for real-world multimodal workloads like voice assistants with visual context.
total_cost = (text_tokens / 1M) × text_rate + (image_tokens / 1M) × image_rate + audio_minutes × audio_rate
Most vision-enabled LLMs convert images into an equivalent token count internally, so they can be billed under the same per-token pricing schedule as text.
This calculator treats each modality as a single blended rate; for detailed input/output split pricing use the Cost per Query Calculator alongside this one.
Set audio minutes to 0 — the audio cost term will simply drop out of the total.