Billing Description (MiaoSearch and MiaoRead)

更新时间:
复制 MD 格式

This document describes the billing for MiaoSearch and MiaoRead.

Billing Overview

  1. MiaoSearch billing consists of two parts: online and offline.

  • Online: The number of Large Language Model (LLM) tokens consumed for Q&A search and pure search.

  • Offline processes include file and index storage, file pre-processing, and vector indexing of files for various modalities.

  1. MiaoRead billing is based on the number of tokens consumed by API calls.

Activation & Product Addresses

MiaoSearch Billing Description

Online Billing Items

On the SaaS page, select either Q&A search or pure search. The system bills based on the selected model and the following rules:

Billing Item

Sub-Billing Item

Billing

Quanmiao-max (Thinking/Non-thinking)

quanmiao-max

0.06 CNY/1,000 Tokens

Quanmiao-Plus (Thinking/Non-thinking)

quanmiao-Plus

0.002 CNY/1,000 Tokens

Offline Billing Items

Billing Item

Sub-Billing Item

Billing

Selection Description

Index Storage (Required, choose one sub-billing item)

Index Storage - Performance Priority

0.0118 CNY/hour × GB

Fast retrieval speed. Suitable for knowledge bases ≥1 TB with many concurrent online users.

Index Storage - Cost Priority

0.0066 CNY/hour × GB

Slow retrieval speed. Suitable for knowledge bases <1 TB with few concurrent online users.

Multimodal File Storage (Optional, select as needed)

Multimodal File Storage

0.0003 CNY/hour × GB

If you only have a text knowledge base, and it is stored in Alibaba Cloud OSS or another cloud service that supports reading, you can skip this item;

If your knowledge base includes videos, this item is required, because the system automatically stores and bills for extracted frames during index building.

File Format Conversion and Processing (Optional, select as needed)

File Format Conversion and Processing

1.20 CNY/hour

Convert audio and video in your knowledge base to text using Automatic Speech Recognition (ASR) for retrieval and recall.

Vector Building for Text and Multimodal Files (Required)

quanmiao-clip

0.0004 CNY/1,000 Tokens

Vectorize (Embedding) text and multimodal files in your knowledge base.

MiaoRead Billing Description

When calling MiaoRead APIs for generation, extraction, and Q&A, specify the model (quanmiao-max or quanmiao-plus) in the ModelName field. Billing is based on the actual token consumption of the selected model.

Billing Item

Billing

quanmiao-max

0.06 CNY/1,000 Tokens

quanmiao-Plus

0.002 CNY/1,000 Tokens

API Call Best Practices

Bill Query

You can click Bill Details to view your bill.