Model Comparison

Qwen3 VL 235B A22B Thinking vs Step3-VL-10BWhich is better in 2026?

Qwen3 VL 235B A22B Thinking significantly outperforms across most benchmarks.

Verdict: Qwen3 VL 235B A22B Thinking vs Step3-VL-10B — which is better?

Qwen3 VL 235B A22B Thinking (by Alibaba Cloud / Qwen Team) and Step3-VL-10B (by StepFun) are two of the AI models people compare most. Here is how they stack up on benchmarks, price and capabilities, and which one to pick in 2026.

Qwen3 VL 235B A22B Thinking outperforms in 2 benchmarks (AIME 2025, MathVision), while Step3-VL-10B is better at 0 benchmarks. Qwen3 VL 235B A22B Thinking significantly outperforms across most benchmarks.

Choose Qwen3 VL 235B A22B Thinking if…

  • you want the strongest raw capability — it leads on 2 of 2 shared benchmarks

Choose Step3-VL-10B if…

  • you want the most recent training data — it shipped Jan 2026

Performance Benchmarks

Comparative analysis across standard metrics

2 benchmarks

Qwen3 VL 235B A22B Thinking outperforms in 2 benchmarks (AIME 2025, MathVision), while Step3-VL-10B is better at 0 benchmarks.

Qwen3 VL 235B A22B Thinking significantly outperforms across most benchmarks.

Wed Jun 24 2026 • llm-stats.com

Arena Performance

Human preference votes

Model Size

Parameter count comparison

226.0B diff

Qwen3 VL 235B A22B Thinking has 226.0B more parameters than Step3-VL-10B, making it 2260.0% larger.

Alibaba Cloud / Qwen Team
Qwen3 VL 235B A22B Thinking
236.0Bparameters
StepFun
Step3-VL-10B
10.0Bparameters
236.0B
Qwen3 VL 235B A22B Thinking
10.0B
Step3-VL-10B

Context Window

Maximum input and output token capacity

Only Qwen3 VL 235B A22B Thinking specifies input context (262,144 tokens). Only Qwen3 VL 235B A22B Thinking specifies output context (262,144 tokens).

Alibaba Cloud / Qwen Team
Qwen3 VL 235B A22B Thinking
Input262,144 tokens
Output262,144 tokens
StepFun
Step3-VL-10B
Input- tokens
Output- tokens
Wed Jun 24 2026 • llm-stats.com

Input Capabilities

Supported data types and modalities

Both Qwen3 VL 235B A22B Thinking and Step3-VL-10B support multimodal inputs.

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

Qwen3 VL 235B A22B Thinking

Text
Images
Audio
Video

Step3-VL-10B

Text
Images
Audio
Video

License

Usage and distribution terms

Both models are licensed under Apache 2.0.

Both models share the same licensing terms, providing consistent usage rights.

Qwen3 VL 235B A22B Thinking

Apache 2.0

Open weights

Step3-VL-10B

Apache 2.0

Open weights

Release Timeline

When each model was launched

Qwen3 VL 235B A22B Thinking was released on 2025-09-22, while Step3-VL-10B was released on 2026-01-15.

Step3-VL-10B is 4 months newer than Qwen3 VL 235B A22B Thinking.

Qwen3 VL 235B A22B Thinking

Sep 22, 2025

9 months ago

Step3-VL-10B

Jan 15, 2026

5 months ago

3mo 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

Key Takeaways

Larger context window (262,144 tokens)
Higher AIME 2025 score (89.7% vs 87.7%)
Higher MathVision score (74.6% vs 70.8%)

No standout differentiators in the data we have for this pair.

Detailed Comparison

AI Model Comparison Table
Feature
Alibaba Cloud / Qwen Team
Qwen3 VL 235B A22B Thinking
StepFun
Step3-VL-10B

FAQ

Common questions about Qwen3 VL 235B A22B Thinking vs Step3-VL-10B.

Which is better, Qwen3 VL 235B A22B Thinking or Step3-VL-10B?

Qwen3 VL 235B A22B Thinking significantly outperforms across most benchmarks. Qwen3 VL 235B A22B Thinking is made by Alibaba Cloud / Qwen Team and Step3-VL-10B is made by StepFun. The best choice depends on your use case — compare their benchmark scores, pricing, and capabilities above.

How does Qwen3 VL 235B A22B Thinking compare to Step3-VL-10B in benchmarks?

Qwen3 VL 235B A22B Thinking scores ZebraLogic: 97.3%, DocVQAtest: 96.5%, ScreenSpot: 95.4%, CountBench: 93.7%, MMLU-Redux: 93.7%. Step3-VL-10B scores MMBench: 91.8%, AIME 2025: 87.7%, MathVista: 84.0%, MMMU: 78.1%, MathVision: 70.8%.

What are the context window sizes for Qwen3 VL 235B A22B Thinking and Step3-VL-10B?

Qwen3 VL 235B A22B Thinking supports 262K tokens and Step3-VL-10B supports an unknown number of tokens. A larger context window lets you process longer documents, conversations, or codebases in a single request.

Who makes Qwen3 VL 235B A22B Thinking and Step3-VL-10B?

Qwen3 VL 235B A22B Thinking is developed by Alibaba Cloud / Qwen Team and Step3-VL-10B is developed by StepFun.