Benchgen
Models/alibaba/

Qwen3 VL 235B A22B Thinking

DraftPublic
Model CardTraining jobsEvaluations

Model Details

Qwen3 VL 235B A22B Thinking

Quick answer: Qwen3 VL 235B A22B Thinking is Alibaba's Feb 2026 model scoring 83.8% MMLU-Pro. Apache 2.0. Alibaba's 235B A22B VL thinking MoE.

What Qwen3 VL 235B A22B Thinking Is

Qwen3 VL 235B A22B Thinking is Alibaba's Feb 2026 release scoring 83.8% MMLU-Pro. Alibaba's 235B A22B VL thinking MoE.

Specifications

FieldValue
OrganizationAlibaba
LicenseApache 2.0
Release dateFeb 2026
ModalityText only

Public Benchmark Scores

BenchmarkScoreSourceDate
MMLU-Pro83.8%Benchgen evaluationFeb-2026

Pricing

Open weights — self-host at no cost.

Qwen3 VL 235B A22B Thinking vs Alternatives

ModelMMLU-ProLicense
Qwen3 VL 235B A22B Thinking83.8%Apache 2.0

Check the MMLU-Pro leaderboard for current comparisons.

Frequently Asked Questions

What is Qwen3 VL 235B A22B Thinking? Alibaba's Feb 2026 model scoring 83.8% MMLU-Pro. Apache 2.0.

Scores from Benchgen evaluations. Last updated 2026-07-24.

Benchmark Leaderboards

This model isn’t on any benchmark leaderboard yet.