The AI arena is free today

Open Superagent

Qwen2.5 VL 72B Instruct vs Qwen3 VL 30B A3B Thinking

Qwen3 VL 30B A3B Thinking leads the LLM Stats Score 18.3 to 12.7.

Alibaba Cloud / Qwen Team · Alibaba Cloud / Qwen Team · Updated for 2026

Which is better?

Qwen3 VL 30B A3B Thinking leads the overall LLM Stats Score 18.3 to 12.7, ranking #212 overall.

In the 15 individual benchmarks reported for both models, Qwen3 VL 30B A3B Thinking wins 12; this is a narrower head-to-head signal than the composite indexes.

Based on current LLM Stats indexes, shared benchmarks, pricing, and model metadata for 2026.

Choose Qwen2.5 VL 72B Instruct

  • you are already invested in the Alibaba Cloud / Qwen Team ecosystem

Choose Qwen3 VL 30B A3B Thinking

  • overall performance matters — it scores 18.3 and ranks #212 on LLM Stats
  • your work emphasizes reasoning — it leads those capability indexes
  • you value its reported benchmark strengths — it wins 12 of 15 exact shared results
  • you want the most recent training data — it shipped Sep 2025

At a glance

The differences that matter most.

Qwen2.5 VL 72B Instruct
Alibaba Cloud / Qwen Team
Qwen3 VL 30B A3B Thinking
Alibaba Cloud / Qwen Team
Core performance indexes
LLM Stats Score
12.7
#249
18.3
#212
Reasoning
12.8
#240
19.5
#198
Agents
4.7
#152
7.2
#136
Cost, coverage & limits
Benchmark wins
3 of 15
12 of 15
Input price
— / M
$0.20 / M
Output price
— / M
$0.99 / M
Context window
—
131,072

Capability indexes

Additional strengths measured across groups of related public benchmarks

3 shared
Index
Qwen2.5 VL 72B Instruct
Qwen3 VL 30B A3B Thinking
Math
7.6#263
22.3#138
Vision
9.2#130
12.2#104
Multimodal
12.1#106
15.0#90
Conservative TrueSkill rating · higher is betterHow scores work

Individual benchmarks

30 reported for Qwen2.5 VL 72B Instruct · 50 for Qwen3 VL 30B A3B Thinking

15 shared

Qwen2.5 VL 72B Instruct outperforms in 3 benchmarks (AI2D, CC-OCR, OCRBench), while Qwen3 VL 30B A3B Thinking is better at 12 benchmarks (Hallusion Bench, LVBench, MathVision, MathVista-Mini, MLVU-M, MMMU-Pro, MMStar, MVBench, OCRBench-V2 (en), OSWorld, ScreenSpot, ScreenSpot Pro).

Qwen3 VL 30B A3B Thinking significantly outperforms across most benchmarks.

Fri Sep 11 2026 • llm-stats.com

Human preference

Blind head-to-head votes and playground preference scores

Model Size

Parameter count comparison

41.0B diff

Qwen2.5 VL 72B Instruct has 41.0B more parameters than Qwen3 VL 30B A3B Thinking, making it 132.3% larger.

Alibaba Cloud / Qwen Team
Qwen2.5 VL 72B Instruct
72.0Bparameters
Alibaba Cloud / Qwen Team
Qwen3 VL 30B A3B Thinking
31.0Bparameters
72.0B
Qwen2.5 VL 72B Instruct
31.0B
Qwen3 VL 30B A3B Thinking

Context Window

Maximum input and output token capacity

Only Qwen3 VL 30B A3B Thinking specifies input context (131,072 tokens). Only Qwen3 VL 30B A3B Thinking specifies output context (32,768 tokens).

Alibaba Cloud / Qwen Team
Qwen2.5 VL 72B Instruct
Input- tokens
Output- tokens
Alibaba Cloud / Qwen Team
Qwen3 VL 30B A3B Thinking
Input131,072 tokens
Output32,768 tokens
Fri Sep 11 2026 • llm-stats.com

Input capabilities

Documented input modalities across available providers

Both Qwen2.5 VL 72B Instruct and Qwen3 VL 30B A3B Thinking support multimodal inputs.

They are both capable of processing various types of data, offering versatility in application.

Qwen2.5 VL 72B Instruct

Text
Images
Audio
Video

Qwen3 VL 30B A3B Thinking

Text
Images
Audio
Video

License

Usage and distribution terms

Qwen2.5 VL 72B Instruct is licensed under tongyi-qianwen, while Qwen3 VL 30B A3B Thinking uses Apache 2.0.

License differences may affect how you can use these models in commercial or open-source projects.

Qwen2.5 VL 72B Instruct

tongyi-qianwen

Open weights

Qwen3 VL 30B A3B Thinking

Apache 2.0

Open weights

Release Timeline

When each model was launched

Qwen2.5 VL 72B Instruct was released on 2025-01-26, while Qwen3 VL 30B A3B Thinking was released on 2025-09-22.

Qwen3 VL 30B A3B Thinking is 8 months newer than Qwen2.5 VL 72B Instruct.

Qwen2.5 VL 72B Instruct

Jan 26, 2025

1.6 years ago

Qwen3 VL 30B A3B Thinking

Sep 22, 2025

11 months ago

7mo newer

Knowledge Cutoff

When training data ends

Neither model specifies a knowledge cutoff date.

Unable to compare the recency of their training data.

No cutoff dates available

Outputs Comparison

Notice missing or incorrect data?Start an Issue discussion→

Judge for yourself.

Run your own prompts against Qwen2.5 VL 72B Instruct and Qwen3 VL 30B A3B Thinking side-by-side, then vote on the output you prefer.

Qwen2.5 VL 72B Instruct
✓ Preferred
Qwen3 VL 30B A3B Thinking
Open in Playground

FAQ

Common questions about Qwen2.5 VL 72B Instruct vs Qwen3 VL 30B A3B Thinking.

Which is better, Qwen2.5 VL 72B Instruct or Qwen3 VL 30B A3B Thinking?

Qwen3 VL 30B A3B Thinking leads the LLM Stats Score 18.3 to 12.7. Qwen2.5 VL 72B Instruct is made by Alibaba Cloud / Qwen Team and Qwen3 VL 30B A3B Thinking is made by Alibaba Cloud / Qwen Team. The best choice depends on your use case — compare their capability indexes, individual benchmarks, pricing, and limits above.

How does Qwen2.5 VL 72B Instruct compare to Qwen3 VL 30B A3B Thinking in benchmarks?

Qwen2.5 VL 72B Instruct scores DocVQA: 96.4%, Android Control Low_EM: 93.7%, ChartQA: 89.5%, OCRBench: 88.5%, AI2D: 88.4%. Qwen3 VL 30B A3B Thinking scores DocVQAtest: 95.0%, ScreenSpot: 94.7%, MMLU-Redux: 90.9%, MMBench-V1.1: 88.9%, MMLU: 87.6%.

What are the context window sizes for Qwen2.5 VL 72B Instruct and Qwen3 VL 30B A3B Thinking?

Qwen2.5 VL 72B Instruct supports an unknown number of tokens and Qwen3 VL 30B A3B Thinking supports 131K tokens. A larger context window lets you process longer documents, conversations, or codebases in a single request.

What are the main differences between Qwen2.5 VL 72B Instruct and Qwen3 VL 30B A3B Thinking?

Key differences include LLM Stats Score (12.7 vs 18.3), licensing (tongyi-qianwen vs Apache 2.0). See the full comparison above for benchmark-by-benchmark results.

Related comparisons