Skip to content
llmwaves

Qwen3 VL 235B A22B Instruct

Qwen · released Sep 23, 2025

Open weightsTool useStructured outputVision

Qwen3-VL-235B-A22B Instruct is an open-weight multimodal model that unifies strong text generation with visual understanding across images and video. The Instruct model targets general vision-language use (VQA, document parsing, chart/table...

Compare this modelSource listing

Specification

Context window
262K
Max output
33K
Knowledge cutoff
Mar 31, 2025
Parameters
235.7B
Licence
apache-2.0
Serving providers
5
Moderated
No
Uptime
99.6%

Intelligence

14.4

47th percentile

Coding

—

Coding Index

Agentic

—

Agentic Index

Output speed

32 t/s

Median across providers

Latency

715ms

Time to first token

Cost per task

$0.06

Estimated

Benchmarks

Where the score comes from

The Intelligence Index is a composite. These are the underlying evaluations this model was actually measured on.

Evaluation scores

Percentage correct · higher is better

  • MMLU-Pro
    82.3%
  • GPQA Diamond
    71.2%
  • AIME 2025
    70.7%
  • LiveCodeBench
    59.4%
  • IFBench
    42.7%
  • SciCode
    35.9%
  • τ²-bench (Telecom)
    35.1%
  • AA-LCR (long context)
    32.0%
  • Terminal-Bench Hard
    6.8%
  • Humanity's Last Exam
    6.6%

An evaluation missing from this list was not run for this model — it is not a zero.

View as table
EvaluationScore
MMLU-Pro82.3%
GPQA Diamond71.2%
AIME 202570.7%
LiveCodeBench59.4%
IFBench42.7%
SciCode35.9%
τ²-bench (Telecom)35.1%
AA-LCR (long context)32.0%
Terminal-Bench Hard6.8%
Humanity's Last Exam6.6%

Against its peers

Intelligence Index · this model highlighted, nearest peers in grey

Peers are the models sitting closest on the Intelligence Index — the set you would realistically choose between.

View as table
ModelIntelligence
Nemotron 3 Nano Omni (free)15.0
GPT-4.1 Mini14.8
Mistral Medium 3.114.7
Solar Pro 314.5
Llama 4 Maverick14.5
Qwen3 VL 235B A22B Instruct14.4
Ling-2.6-flash14.2
Gemini 2.5 Flash14.2

Percentile among all indexed models

Intelligence47th
Mathematics65th
Terminal work35th

Pricing

What it costs to run

List prices per million tokens, plus what one representative task works out to.

List price

Input / 1M tokens
$0.21
Output / 1M tokens
$1.9
Cached input / 1M
$0.1
Blended 3:1
$0.632

One task, estimated

$0.06

Input tokens
50,000
Output tokens
25,000
Profile
Standard

Estimated from list pricing: 50K input tokens plus 80K output tokens for reasoning models (25K for non-reasoning).

Serving providers

Speed and latency figures are medians across these providers, so a widely-served model reports a blend rather than any single endpoint.

Same creator

More from Qwen

Qwen3.8 Max

Intelligence
58.1
Speed
38 t/s
Cost per task
$0.58

Qwen3.7 Max

Intelligence
46.7
Speed
47 t/s
Cost per task
$0.43

Qwen3.6 Max Preview

Intelligence
41.1
Speed
47 t/s
Cost per task
$0.54

Qwen3.6 Plus

Intelligence
40.5
Speed
39 t/s
Cost per task
$0.17

Qwen3.7 Plus

Intelligence
39.4
Speed
12 t/s
Cost per task
$0.12

Qwen3.6 27B

Intelligence
37.7
Speed
77 t/s
Cost per task
$0.32