ModelRushText

Qwen Audio 3.0 Realtime Flash API

Qwen Audio 3.0 Realtime Flash runs realtime sessions from streamed audio and text through the ModelRush API and returns a streamed session. Requests run synchronously and the response carries the result. It is callable in 1 execution region, from $4.5386 per 1M input tokens.
Real-time multimodal
Current rate
$4.5386/ 1M tokens
Try it in the PlaygroundLet an agent integrate this model

Input

sync
Estimated run$4.5386
Sign in to generate

Output preview

This is a model capability preview. Sign in to submit a live request and inspect the generated result here.
token5s

Model information

Model ID
modelrush/qwen-audio-3.0-realtime-flash
Integration status
Inputs
Text
Output
Text
Billing unit
Per 1M tokens
Execution mode
Synchronous. The response carries the result.
Available regions
China — Beijing
Last updated
Sep 8, 2026

Best for

Product featuresAutomated workflowsAgent tools

Current customer pricing

Each row identifies the Operation, Region, price variant, and immutable price version used for execution.
ModelCapabilityExecution RegionPrice variantCustomer price
Qwen Audio 3.0 Realtime Flash · modelrush/qwen-audio-3.0-realtime-flashRealtimecn-beijingStandardInput $4.5386 · Output $15.1272 / 1M tokens

Pricing and documentation

Qwen Audio 3.0 Realtime Flash API pricingAsynchronous jobs and webhooks documentation

Other recommended models

LLM

DeepSeek V4 Flash 0731

Fast, cost-efficient reasoning and text generation
Starting price
$0.424/ 1M tokens
LLM

DeepSeek V4 Pro

Reasoning model · released in 4 regions
Starting pricePricing pending
LLM

GLM 5.1

Reasoning model · released in 5 regions
Starting pricePricing pending
ModelRushOne integration, intelligent routing, transparent billing. Model infrastructure for developers and agents.
ProductModelsPricingDashboardBlog
DevelopersDocsAgent integrationDIY AI Girlfriend
© 2026 ModelRushAll systems operational
Privacy & TermsAcceptable UseData RetentionData Deletion Policy