impulse

Qwen3 VL 235B A22B Instruct

Open weights

Qwen3-VL-235B-A22B Instruct is an open-weight multimodal model that unifies strong text generation with visual understanding across images and video.

API id qwen3-vl-235b-a22b-instructReleased 23 Sep 2025 (Sep 2025)
Input / 1M
$0.40
at Alibaba DashScope
Output / 1M
$1.60
at Alibaba DashScope
Context window
131K
tokens
Max output
33K
tokens per reply
Hosts
9
9 listings
Released
Sep 2025
23 Sep 2025

Pricing

List prices at Alibaba DashScope, in US dollars per million tokens.

TierInput / 1MOutput / 1M
Standard$0.40$1.60

What it costs a month

At Alibaba DashScope list prices, 30 days.

  • Chat assistant5,000 requests a day, 2K in / 0.5K out$240
  • RAG over documents2,000 requests a day, 8K in / 0.6K out$250
  • Coding agent1,000 requests a day, 40K in / 2K out$576
  • Overnight batch20,000 requests a day, 4K in / 0.4K out$1,344
Use your own numbers

Where to run it

9 hosts serve Qwen3 VL 235B A22B Instruct. The comparison is on blended price, 3 parts input to 1 part output.

HostIn / out per 1M
qwen3-vl-235b-a22b-instruct
$0.40 / $1.60
qwen3-vl-235b-a22b-instruct
$0.40 / $1.60
Qwen CloudLab API
qwen3-vl-235b-a22b-instruct
$0.40 / $1.60
DeepInfraInference
Qwen3-VL-235B-A22B-Instruct
$0.20 / $0.88
−47% vs headline
Fireworks AIInference
qwen3-vl-235b-a22b-instruct
$0.22 / $0.88
−45% vs headline
Qwen3-VL-235B-A22B-Instruct
$0.26 / $1.15
−31% vs headline
GMI CloudInference
Qwen3-VL-235B-A22B-Instruct-FP8
$0.30 / $1.40
−18% vs headline
Novita AIInference
qwen3-vl-235b-a22b-instruct
$0.30 / $1.50
−14% vs headline
Mantle
qwen.qwen3-vl-235b-a22b-instruct
$0.53 / $2.66
+52% vs headline

Capabilities

Core

  • Streaming
  • Function calling
  • JSON mode
  • Structured outputs

Multimodal

  • Vision
  • Image input
  • Audio
  • Image generation
  • Video

Advanced

  • Reasoning
  • Tool calling
  • MCP
  • Prompt caching
  • Batch API
  • Embeddings
Takes in
TextImages
Produces
Text
Open weights
License: Apache-2.0
Runs in about 565.6 GB of GPU memory
Runtimes: Ollama, SGLang, llama.cpp, vLLM

Cheaper alternatives

Current models that cost less and keep most of what this one can do: tools, vision, reasoning, structured output, caching and enough context.

Same model, cheaper host

  • DeepInfra
    Qwen3-VL-235B-A22B-Instruct
    $0.20 / $0.88
    47% less
  • Fireworks AI
    qwen3-vl-235b-a22b-instruct
    $0.22 / $0.88
    45% less
  • Inference APIs
    Qwen3-VL-235B-A22B-Instruct
    $0.26 / $1.15
    31% less
  • GMI Cloud
    Qwen3-VL-235B-A22B-Instruct-FP8
    $0.30 / $1.40
    18% less
  • Novita AI
    qwen3-vl-235b-a22b-instruct
    $0.30 / $1.50
    14% less

Smaller Alibaba Qwen models

No cheaper Alibaba Qwen model keeps its key capabilities.

Other frontier labs

  • GPT-5.4 Nano
    OpenAI · on OpenAI
    $0.20 / $1.25
    Has every key capability
    34% less
  • Claude Haiku 5.5
    Anthropic · on Anthropic
    $0.10 / $0.50
    Has every key capability
    71% less
  • GPT-6 Luna
    OpenAI · on OpenAI
    $0.10 / $0.50
    Has every key capability
    71% less
  • GPT-4o Mini
    OpenAI · on OpenAI
    $0.15 / $0.60
    Has every key capability
    63% less

Open-weight picks

  • Kimi K2.7 Code
    Moonshot AI · on CLF AI Gateway
    $0.38 / $1.60
    Has every key capability
    2% less
  • Inkling Small
    Independent · on DeepInfra
    $0.45 / $1.20
    Has every key capability
    9% less
  • DeepSeek Flash
    DeepSeek · on DeepSeek
    $0.30 / $1.20
    Has every key capability
    25% less
  • Qwen3.7 Plus
    Alibaba Qwen · on Replicate
    $0.276 / $1.10
    Has every key capability
    31% less

Call it

Using Alibaba DashScope's API.

Python
from litellm import completion

response = completion(
    model="dashscope/qwen3-vl-235b-a22b-instruct",
    messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)

Quick answers

How much does Qwen3 VL 235B A22B Instruct cost?

On Alibaba DashScope, Qwen3 VL 235B A22B Instruct costs $0.40 per million input tokens and $1.60 per million output tokens. The cheapest host is DeepInfra, at $0.20 per million input tokens and $0.88 per million output tokens.

Which APIs offer Qwen3 VL 235B A22B Instruct?

Qwen3 VL 235B A22B Instruct runs on 9 APIs: Alibaba DashScope, Qianwen AI Platform, Qwen Cloud, DeepInfra, Fireworks AI, Inference APIs, GMI Cloud, Novita AI and Amazon Bedrock.

What is Qwen3 VL 235B A22B Instruct's context window?

131K tokens, with up to 33K tokens in one reply.

Is Qwen3 VL 235B A22B Instruct being retired?

No retirement date is scheduled for it on any API we track.

What are cheaper alternatives to Qwen3 VL 235B A22B Instruct?

Models that cost less and keep most of what it can do: GPT-5.4 Nano on OpenAI (34% less), Claude Haiku 5.5 on Anthropic (71% less) and Kimi K2.7 Code on CLF AI Gateway (2% less). Qwen3 VL 235B A22B Instruct itself costs 47% less on DeepInfra. Savings are on a mix of 3 input tokens to 1 output token.

When was Qwen3 VL 235B A22B Instruct released?

Alibaba Qwen released it on 23 Sep 2025.

What is Qwen3 VL 235B A22B Instruct's API model id?

Use qwen3-vl-235b-a22b-instruct with Alibaba DashScope. Other hosts use their own ids: Qwen3-VL-235B-A22B-Instruct on DeepInfra, Qwen3-VL-235B-A22B-Instruct on Inference APIs, Qwen3-VL-235B-A22B-Instruct-FP8 on GMI Cloud, qwen.qwen3-vl-235b-a22b-instruct on Amazon Bedrock.

Sources

Raw catalog entry for qwen3-vl-235b-a22b-instruct on Alibaba DashScope
{
  "input_cost_per_token": 4e-7,
  "litellm_provider": "dashscope",
  "max_input_tokens": 131072,
  "max_output_tokens": 32768,
  "mode": "chat",
  "output_cost_per_token": 0.0000016,
  "source": "https://www.alibabacloud.com/help/en/model-studio/model-pricing",
  "supports_function_calling": true,
  "supports_tool_choice": true,
  "supports_vision": true
}