Agent Metering and Billing

Last Updated on : 2025-04-02 06:02:06download

Overview

This topic describes the various billing items and related costs of the AI agent.

Billing items

The agent’s billing items currently include model fees and smart voice fees.

Billing cycle

Each billing item is charged daily, and the fees are settled based on actual usage. Bills are typically generated one day after the current billing cycle ends, with the exact timing determined by the system.

Model fees

Available model types

For more information about how to differentiate model types, see Available Models.

Billing formula

Model services are charged based on the model’s token usage. The billing formula is as follows:

Model fees = Token usage × Unit price of tokens

In a large language model, a token is the basic unit of text processing. The model usually breaks down the input text into a series of tokens and then processes and analyzes these tokens. Tokens can be words, characters, subword fragments, or other text segments. The specific segmentation is determined by the model’s tokenization algorithm. Therefore, token calculation and processing methods might vary depending on the model’s architecture and design.

Unit price

Model name	Billing item	Unit price (per million tokens)
Qwen	Qwen-Max input	¥2.40
Qwen	Qwen-Max output	¥9.60
Doubao	Doubao-Pro-32k input	¥0.80
Doubao	Doubao-Pro-32k output	¥2.00
DeepSeek	Deepseek-Chat input	¥2.00
DeepSeek	Deepseek-Chat output	¥8.00
ChatGPT	GPT-4o input	$2.50
	GPT-4o output	$10.00
	GPT-4o-mini input	$0.15
	GPT-4o-mini output	$0.60
Gemini	Gemini-1.5-pro input	$1.25
	Gemini-1.5-pro output	$2.50
	Gemini-2.0-flash input	$0.10
	Gemini-2.0-flash output	$0.40
Mistral	Mistral Large input	-
Mistral	Mistral Large output	-
Claude	Claude 3.5 Haiku input	$0.80
	Claude 3.5 Haiku output	$4.00
	Claude 3.7 Sonnet input	$3.00
	Claude 3.7 Sonnet output	$15.00
Nova	Nova Pro input	-
Nova	Nova Pro output	-

Smart voice fees

Billing formula

Smart voice service consists of voice input (ASR) and voice output (TTS). The billing formula is as follows:

Smart voice fee = ASR unit price × input audio duration + TTS unit price × output audio character count

Automatic speech recognition (ASR): A technology that can recognize and understand natural human audio language input, converting audio into text by analyzing and processing speech signals.
Text to speech (TTS): A technology that converts text into speech output, simulating human reading aloud to transform written information into audio.

Unit price

ASR vendor	Unit price (per hour)
AISpeech	¥5.00
Google	$1.44

TTS vendor	Unit price (per thousand characters)
AISpeech	¥0.20
Volcano Ark	¥0.30
Azure	$0.015

Prev DocVariables Management

Next DocAI Product Commands