QwenReleased on Sep 22, 2025

Qwen3 VL 235B A22B Thinking: Benchmarks, Pricing & Size

Qwen3 VL 235B A22B Thinking is a language model from Qwen, released in September 2025, with multimodal input.

Qwen3-VL-235B-A22B-Thinking is the most powerful vision-language model in the Qwen series, featuring 236B parameters with MoE architecture for reasoning-enhanced multimodal understanding. Key capabilities include: Visual Agent (operates

Input
TextImageVideo
Output
Text
Speed
—
Cost
$0.79/ 1M · 8:1 in:out
$0.45 in · $3.49 out

Qwen3 VL 235B A22B Thinking pricing

Providers

Qwen3 VL 235B A22B Thinking starts at $0.450 per million input tokens and $3.49 per million output tokens via DeepInfra. See all 2 providers below with their per-token pricing, latency, throughput, and modality support.

ProviderInput $/MOutput $/MMax InputMax OutputLatency p95 sThroughput P95QuantInputOutput
DeepInfra logoDeepInfra
$0.450$3.49262.1K262.1K
—
—
fp8
Novita logoNovita
$0.980$3.95131.1K32.8K
—
—
bf16
Loading chart...
Loading chart...
Loading chart...

Qwen3 VL 235B A22B Thinking model size

Qwen3 VL 235B A22B Thinking has 236 billion parameters. See how it compares to other models in the same parameter range.

Parameters
236BMoE
Frontier (200B+)
236B
1B7B70B405B

Qwen3 VL 235B A22B Thinking API

API access coming soon

Qwen3 VL 235B A22B Thinking will be available through our gateway shortly.

Qwen3 VL 235B A22B Thinking examples

Recent arena outputs from Qwen3 VL 235B A22B Thinking, picked from the highest-ranked matchups.

Qwen3 VL 235B A22B Thinking license

Qwen3 VL 235B A22B Thinking is released under the Apache 2.0 license, which permits commercial use, has 236.0B parameters.

License
Apache 2.0
Commercial use allowed
Parameters
236.0B

Apache License 2.0 - allows commercial use

FAQ

Common questions about Qwen3 VL 235B A22B Thinking.

When was Qwen3 VL 235B A22B Thinking released?

Qwen3 VL 235B A22B Thinking was released on September 22, 2025 by Qwen. This is the official Qwen3 VL 235B A22B Thinking release date tracked on LLM Stats.

Is Qwen3 VL 235B A22B Thinking available via API?

Yes, Qwen3 VL 235B A22B Thinking is available via API. See the official documentation for authentication and endpoint details. It is served by 2 providers tracked on LLM Stats.

How big is Qwen3 VL 235B A22B Thinking?

Qwen3 VL 235B A22B Thinking has 236 billion parameters. It ships as an open-weight model, so you can download and run it on your own hardware.

Who created Qwen3 VL 235B A22B Thinking?

Qwen3 VL 235B A22B Thinking was created by Qwen.

What is the license for Qwen3 VL 235B A22B Thinking?

Qwen3 VL 235B A22B Thinking is released under the Apache 2.0 license. This is an open-source / open-weight license that permits self-hosting.

Is Qwen3 VL 235B A22B Thinking multimodal?

Yes, Qwen3 VL 235B A22B Thinking is multimodal and can accept both text and images as input.

Where can I use Qwen3 VL 235B A22B Thinking?

Qwen3 VL 235B A22B Thinking is available through 2 providers including DeepInfra, Novita.

Where is the Qwen3 VL 235B A22B Thinking paper or technical report?

Qwen3 VL 235B A22B Thinking has a paper or technical report available at https://arxiv.org/abs/2505.09388. Use that source for architecture, training, release and evaluation details.