Model Comparison
Qwen3-235B-A22B-Thinking-2507 vs Qwen3 VL 30B A3B InstructWhich is better in 2026?
Qwen3-235B-A22B-Thinking-2507 significantly outperforms across most benchmarks. Qwen3 VL 30B A3B Instruct is 3.0x cheaper per token.
Verdict: Qwen3-235B-A22B-Thinking-2507 vs Qwen3 VL 30B A3B Instruct — which is better?
Qwen3-235B-A22B-Thinking-2507 (by Alibaba Cloud / Qwen Team) and Qwen3 VL 30B A3B Instruct (by Alibaba Cloud / Qwen Team) 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.
Qwen3-235B-A22B-Thinking-2507 outperforms in 17 benchmarks (AIME 2025, Arena-Hard v2, BFCL-v3, Creative Writing v3, GPQA, HMMT25, IFEval, Include, LiveBench 20241125, LiveCodeBench v6, MMLU-Pro, MMLU-ProX, MMLU-Redux, Multi-IF, PolyMATH, SuperGPQA, WritingBench), while Qwen3 VL 30B A3B Instruct is better at 0 benchmarks. Qwen3-235B-A22B-Thinking-2507 significantly outperforms across most benchmarks.
On price, Qwen3 VL 30B A3B Instruct is roughly 3.0x cheaper per token on a blended 3:1 input/output basis, which adds up quickly at production volume.
Qwen3-235B-A22B-Thinking-2507 also accepts a larger context window (262,144 input tokens), making it the stronger choice for long documents and large codebases.
Choose Qwen3-235B-A22B-Thinking-2507 if…
- you want the strongest raw capability — it leads on 17 of 17 shared benchmarks
- you process long inputs — it offers a 262,144 token context window
Choose Qwen3 VL 30B A3B Instruct if…
- cost matters — it's about 3.0x cheaper per token
- you want the most recent training data — it shipped Sep 2025
Performance Benchmarks
Comparative analysis across standard metrics
Qwen3-235B-A22B-Thinking-2507 outperforms in 17 benchmarks (AIME 2025, Arena-Hard v2, BFCL-v3, Creative Writing v3, GPQA, HMMT25, IFEval, Include, LiveBench 20241125, LiveCodeBench v6, MMLU-Pro, MMLU-ProX, MMLU-Redux, Multi-IF, PolyMATH, SuperGPQA, WritingBench), while Qwen3 VL 30B A3B Instruct is better at 0 benchmarks.
Qwen3-235B-A22B-Thinking-2507 significantly outperforms across most benchmarks.
Arena Performance
Human preference votes
Pricing Analysis
Price comparison per million tokens
For input processing, Qwen3-235B-A22B-Thinking-2507 ($0.30/1M tokens) is 1.5x more expensive than Qwen3 VL 30B A3B Instruct ($0.20/1M tokens).
For output processing, Qwen3-235B-A22B-Thinking-2507 ($3.00/1M tokens) is 4.3x more expensive than Qwen3 VL 30B A3B Instruct ($0.70/1M tokens).
In conclusion, Qwen3-235B-A22B-Thinking-2507 is more expensive than Qwen3 VL 30B A3B Instruct.*
* Using a 3:1 ratio of input to output tokens
Model Size
Parameter count comparison
Qwen3-235B-A22B-Thinking-2507 has 204.0B more parameters than Qwen3 VL 30B A3B Instruct, making it 658.1% larger.
Context Window
Maximum input and output token capacity
Qwen3-235B-A22B-Thinking-2507 accepts 262,144 input tokens compared to Qwen3 VL 30B A3B Instruct's 131,072 tokens. Qwen3-235B-A22B-Thinking-2507 can generate longer responses up to 131,072 tokens, while Qwen3 VL 30B A3B Instruct is limited to 32,768 tokens.
Input Capabilities
Supported data types and modalities
Qwen3 VL 30B A3B Instruct supports multimodal inputs, whereas Qwen3-235B-A22B-Thinking-2507 does not.
Qwen3 VL 30B A3B Instruct can handle both text and other forms of data like images, making it suitable for multimodal applications.
Qwen3-235B-A22B-Thinking-2507
Qwen3 VL 30B A3B Instruct
License
Usage and distribution terms
Both models are licensed under Apache 2.0.
Both models share the same licensing terms, providing consistent usage rights.
Apache 2.0
Open weights
Apache 2.0
Open weights
Release Timeline
When each model was launched
Qwen3-235B-A22B-Thinking-2507 was released on 2025-07-25, while Qwen3 VL 30B A3B Instruct was released on 2025-09-22.
Qwen3 VL 30B A3B Instruct is 2 months newer than Qwen3-235B-A22B-Thinking-2507.
Jul 25, 2025
1.0 years ago
Sep 22, 2025
10 months ago
1mo newerKnowledge Cutoff
When training data ends
Neither model specifies a knowledge cutoff date.
Unable to compare the recency of their training data.
Provider Availability
Qwen3-235B-A22B-Thinking-2507 is available from Fireworks, Novita. Qwen3 VL 30B A3B Instruct is available from Novita, DeepInfra.
Qwen3-235B-A22B-Thinking-2507
Qwen3 VL 30B A3B Instruct
Outputs Comparison
Key Takeaways
Qwen3-235B-A22B-Thinking-2507
View detailsAlibaba Cloud / Qwen Team
Qwen3 VL 30B A3B Instruct
View detailsAlibaba Cloud / Qwen Team
Detailed Comparison
Interactive Arena
Judge for yourself.
Run your own prompts against Qwen3-235B-A22B-Thinking-2507 and Qwen3 VL 30B A3B Instruct side-by-side, then vote on the output you prefer.
| Feature | | |
|---|
FAQ
Common questions about Qwen3-235B-A22B-Thinking-2507 vs Qwen3 VL 30B A3B Instruct.