Alibaba Cloud: Qwen3-Omni-Flash

Qwen3 Omni Flash is a multimodal model for understanding information across text, images, and audio. It is useful for assistants that need to discuss visual or spoken material within the same conversation.

Compare models

qwen3-omni-flash

AvailableAlibaba CloudChat

Input / Output: $0.43 / $1.66
Context: 49K
Max output: 16K
Modalities: VisionChat

About Qwen3-Omni-Flash

Qwen3-Omni-Flash is Alibaba's earlier Omni chat model, reading text, images, audio, and video in a single conversation. It can run in a thinking mode that reasons before answering, and in that mode it returns text only; without it, replies can include spoken audio. It is the older and smaller-context member of the Omni chat family, behind Qwen3.5-Omni and Qwen3.8-Omni-Flash, with shorter audio and video limits and fewer output languages.

Where it works well

Understands text, images, audio, and video together, so one request can combine a clip with a written question.
An optional thinking mode lets it reason through a hard question before answering.
Without thinking, it can answer in speech as well as text, using a choice of voices.
Images, audio, and video are passed as content parts in an ordinary chat message.

When to choose another model

Audio output is switched off in thinking mode, so reasoned answers come back as text only.
It handles shorter media than later Omni models: about 20 minutes of audio and 150 seconds of video.
Spoken replies cover 11 languages and 8 dialects, fewer than Qwen3.5-Omni.

Getting started

Create API key
Create a key in Console, then use it with every model on the platform.

Send your first request

Copy the example for your language and run it against the endpoint.

Text to textMessages API

POST/v1/messages

API format:

curl https://api.tokenlab.sh/v1/messages \
  -H "Content-Type: application/json" \
  -H "x-api-key: sk-xxx" \
  -H "anthropic-version: 2023-06-01" \
  -d '{
    "model": "qwen3-omni-flash",
    "max_tokens": 1024,
    "messages": [
      {"role": "user", "content": "Hello!"}
    ]
  }'

Pricing

The Verified price applies to Verified, which costs less on most models. The Official price is the model maker's published price and applies to the more reliable Official route. Auto bills the route that completes the request.

Text Text Output

per 1M tokens

Official price: Input $0.43 / Output $1.66 / Text Input $0.43 / Audio Input $3.81 / Image Input $0.78 / Text Output $1.66 / Audio Output $15.11
Verified price: —
Discount: —

Audio on

per 1M tokens

Official price: Input $0.43 / Output $1.66 / Text Input $0.43 / Audio Input $3.81 / Image Input $0.78 / Text Output $0.00 / Audio Output $15.11
Verified price: —
Discount: —

isMultimodalInput · true

per 1M tokens

Official price: Input $0.43 / Output $1.66 / Text Input $0.43 / Audio Input $3.81 / Image Input $0.78 / Text Output $3.06 / Audio Output $15.11
Verified price: —
Discount: —

Audio on · isMultimodalInput · true

per 1M tokens

Official price: Input $0.43 / Output $1.66 / Text Input $0.43 / Audio Input $3.81 / Image Input $0.78 / Text Output $0.00 / Audio Output $15.11
Verified price: —
Discount: —

	Official priceper 1M tokens	Verified priceper 1M tokens	Discount
Text Text Output	Input $0.43 / Output $1.66 / Text Input $0.43 / Audio Input $3.81 / Image Input $0.78 / Text Output $1.66 / Audio Output $15.11	—	—
Audio on	Input $0.43 / Output $1.66 / Text Input $0.43 / Audio Input $3.81 / Image Input $0.78 / Text Output $0.00 / Audio Output $15.11	—	—
isMultimodalInput · true	Input $0.43 / Output $1.66 / Text Input $0.43 / Audio Input $3.81 / Image Input $0.78 / Text Output $3.06 / Audio Output $15.11	—	—
Audio on · isMultimodalInput · true	Input $0.43 / Output $1.66 / Text Input $0.43 / Audio Input $3.81 / Image Input $0.78 / Text Output $0.00 / Audio Output $15.11	—	—

Usage & activity

Success rate is the share of requests that completed. Latency is how long a full response takes; P95 means 95% of requests finished within that time.

Usage & availability

Last 24 hours

Requests
Success rate
P95 latency
Total tokens

Model performance

Last updated Nov 1, 2026, 6:36 AM UTC

Metrics appear once privacy and data volume thresholds are met.

Data is based on aggregate user requests, excluding status checks.

Open in Console

Open Qwen3-Omni-Flash in Console with a prompt ready to edit or send.

Help me try qwen3-omni-flash with a short message at /v1/messages. Show the reply, latency, and cost.

Use cases

Best for

Vision

Voice-and-image assistants
Let a user speak a question about a photo and get a spoken answer in a chosen voice, without chaining separate speech and vision models.
Short clip analysis
Describe what happens in a clip of up to a couple of minutes, including what is said and shown.
Audio Q&A
Ask questions about a recorded call or voice memo and get a written answer, with thinking on for tricky comparisons.

Prompt examples

Listen to this voice memo and list the three action items with who owns each.

What is the person in this clip doing wrong, and what should they change?

Look at this photo of my router and tell me, out loud, which cable goes where.

This model has conditional pricing. Monthly token totals alone cannot produce a reliable estimate; use the detailed pricing for the request specification, cache, and applicable time window.

FAQ

What can Qwen3-Omni-Flash take as input?

Text, images, audio, and video. All of them can appear in one conversation, so a request can include a spoken question and a picture. Output is text, or text plus audio when thinking mode is off.

Why does thinking mode drop audio?

Alibaba documents that audio output is not supported in thinking mode. If you need a spoken reply, leave thinking off; if you need the model to reason first, expect a written answer and convert it with a text-to-speech model.

How is it different from Qwen3.5-Omni-Flash?

Qwen3.5-Omni-Flash handles longer media, offers more voices and output languages, and has a much larger context. Qwen3-Omni-Flash is the older generation, so use it mainly if your prompts and voices are already tuned for it.

How long can the media be?

About 20 minutes of audio and 150 seconds of video per request, which is shorter than the later Omni models allow. Split longer recordings into parts before sending them.

Which languages can it speak?

Spoken replies cover 11 languages and 8 dialects, fewer than Qwen3.5-Omni. If you need wider spoken coverage, pick a Qwen3.5-Omni model; for text answers the spoken-language list does not apply.

How much does Qwen3-Omni-Flash cost?

On TokenLab, Qwen3-Omni-Flash costs Input $0.43 / Output $1.66 per 1M tokens. The pricing table above shows the full breakdown. Rates depend on the billing unit, specification, and usage. Compare matching conditions in the model's detailed pricing; a single rate does not determine the total cost.

What are the context window and output limit of Qwen3-Omni-Flash?

Qwen3-Omni-Flash accepts up to 49,152 tokens of context and returns up to 16,384 tokens in one response.

Which endpoint should Qwen3-Omni-Flash use?

Use https://api.tokenlab.sh/v1/messages for Qwen3-Omni-Flash. The request example below shows the matching code shape.

Which operations does Qwen3-Omni-Flash support?

Qwen3-Omni-Flash supports Text to text. Select an operation above to see its endpoint and request example.