Alibaba: Qwen3 VL 30B A3B Instruct

Open vision-language MoE model for image and video understanding, documents, and visual agents

Model lab: Alibaba
Model family: qwen

Model resources

Hugging Face

Modalitiestext, image, video → text

Context window262,144

Maximum output32,768

Input · Context ≤ 131072 tokens · Maximum output 32768 tokensUSD 0.16 / 1M tokens

Output · Context ≤ 131072 tokens · Maximum output 32768 tokensUSD 0.56 / 1M tokens

Try now

Quickstart

Use your Moduck API Key, set the API base URL to https://api.moduck.ai/v1, and set the model parameter to alibaba/qwen3-vl-30b-a3b-instruct.

import os
import requests

base_url = "https://api.moduck.ai/v1"
api_key = os.environ["MODUCK_API_KEY"]
model_id = "alibaba/qwen3-vl-30b-a3b-instruct"

response = requests.post(
    f"{base_url}/chat/completions",
    headers={
        "Authorization": f"Bearer {api_key}",
        "Content-Type": "application/json",
    },
    json={
        "model": model_id,
        "messages": [{"role": "user", "content": "Hello!"}],
        "stream": False,
    },
    timeout=60,
)
response.raise_for_status()
print(response.json())

Use the familiar OpenAI-style chat request format to make model integration easier.

Qwen3 VL 30B A3B Instruct API Pricing

Type	Price information
Input · Context ≤ 131072 tokens · Maximum output 32768 tokens	USD 0.16 / 1M tokens
Output · Context ≤ 131072 tokens · Maximum output 32768 tokens	USD 0.56 / 1M tokens

Benchmarks

Intelligence Index · v4.37.9

Coding Index—

Agentic Index—

Performance

Output speed25.34 tokens/s

Time to first token2.84 s

Time to first answer token—

End-to-end response time—

Performance trends

Throughput25.34 tok/sCurrent

Time to first token2.84 sCurrent

End-to-end latency7.3 sCurrent

Throughput

Current25.34 tok/s

Window average24.5 tok/s

Time to first token

Current2.84 s

Window average2.8 s

End-to-end latency

Current7.3 s

Window average7.5 s

Tail latency (P99)

Current16.3 s

Window average16.9 s

Prompt cache hit rate

Current26.4%

Window average24.0%

Error rate

Current0.0%

Window average0.1%

Availability

Uptime100%At least one provider responded

Availability100%Requests actually served

Error rate0%Share of failed requests

Served requests, hour by hour

24 hours agoNow

Qwen3 VL 30B A3B Instruct Compare models

Model	Input price	Output price	Context window	Capabilities
Qwen3 VL 30B A3B InstructThis page	USD 0.16 / 1M tokens · Context ≤ 131072 tokens · Maximum output 32768 tokens	USD 0.56 / 1M tokens · Context ≤ 131072 tokens · Maximum output 32768 tokens	262,144	Text output · Tool calling
Qwen3.8 Flash Next	USD 0.08 / 1M tokens	USD 0.16 / 1M tokens	262,144	Text output · Reasoning · Tool calling
Qwen3.8 Flash	USD 0.09008 / 1M tokens	USD 0.30402 / 1M tokens	1,000,000	Text output · Reasoning · Tool calling
Qwen3.8 27B	USD 0.016 / 1M tokens · Quantization fp4	USD 0.4 / 1M tokens · Quantization fp4	262,144	Text output · Reasoning · Tool calling
Qwen3.8 Max Preview	USD 2 / 1M tokens	USD 6 / 1M tokens	1,000,000	Text output · Reasoning · Tool calling

Model specifications

Input: text, image, video
Output: text
Context window: 262,144
Maximum output: 32,768
Release date: 2025-10-04
Attachments: Yes
Reasoning: No
Tool calling: Yes
Temperature setting: Yes
Open weights: Yes

Frequently asked questions

What is Qwen3 VL 30B A3B Instruct for?

Open vision-language MoE model for image and video understanding, documents, and visual agents

What inputs and outputs does Qwen3 VL 30B A3B Instruct support?

Recorded inputs: text, images, video; outputs: text.

How is Qwen3 VL 30B A3B Instruct priced?

The pricing table shows the current recorded Moduck price. For example: Input · Context ≤ 131072 tokens · Maximum output 32768 tokens USD 0.16 / 1M tokens; Output · Context ≤ 131072 tokens · Maximum output 32768 tokens USD 0.56 / 1M tokens. See the pricing table for all units, variants, and conditions.

What specifications are recorded for Qwen3 VL 30B A3B Instruct?

Current model records list Context window 262,144 tokens; Maximum output 32,768 tokens.

What model parameter does the Quickstart use?

The example uses alibaba/qwen3-vl-30b-a3b-instruct. This value comes from the page identifier and is not confirmed as a callable Moduck public ID.

Can I call this model through the Moduck API now?

This page has no verified Moduck API call for this model. A Quickstart example or reserved ID does not establish access; check the models available to your account.