Qwen

Qwen3 VL 235B A22B Thinking

by Qwentext+image->text2 endpoints
Neura Intelligence Index
43.5/ 100
Rank
#211
Confidence
medium
Benchmarks
7(18% coverage)

Overview

Qwen3-VL-235B-A22B Thinking is a multimodal model that unifies strong text generation with visual understanding across images and video. The Thinking model is optimized for multimodal reasoning in STEM and math....

Capabilities

Text generation
Image understanding
Tool use / Function calling
Structured output
Extended reasoning
Long context
Top-P sampling
Stop sequences
Deterministic seed
Log probabilities

Modalities

Input
TextImage
Output
Text

Technical Specifications

Context Window
131.1K tokens
Max Output
32.8K tokens
Knowledge Cutoff
2025-03-31
Tokenizer
Qwen3
Uptime (24h)
9976.6%
Parameters
236000000000B
License
apache_2_0Open Weights
Country
🇨🇳CN
Release Date
September 22, 202512mo

Supported Parameters

frequency_penaltyinclude_reasoninglogprobsmax_tokenspresence_penaltyreasoningrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_logprobstop_p

Benchmark Performance

general

BenchmarkScoreSource
SimpleQA
44.4
verified

math

BenchmarkScoreSource
AIME 2025
89.7
verified

reasoning

BenchmarkScoreSource
Humanity's Last Exam
13.6
verified

multimodal

BenchmarkScoreSource
MMMU-Pro
69.3
verified
CharXiv Reasoning
66.1
verified
ScreenSpot Pro
61.8
verified

agent

BenchmarkScoreSource
OSWorld
38.1
verified

Performance Over Time

Pricing

Per 1M tokens
Input$0.40
Output$4.00
Added
September 23, 2025
Last synced 9/18/2026