Model Comparison
Qwen3 235B A22B (Non-reasoning)
vs. Qwen3 32B (Reasoning)
Comparing 2 AI models · 12 benchmarks · Alibaba
Recommended Pick
Al Qwen3 32B (Reasoning) 17 metric wins Strongest on: Value, Input price, Output price
Best Value
Qwen3 32B (Reasoning)
100.0 value score
59.2 reasoning / $0.28/1M
Lowest Price
Qwen3 32B (Reasoning)
$0.20/1M input price
Best Reasoning
Qwen3 32B (Reasoning)
59.2 reasoning score
Blends available reasoning benchmarks
Best for Coding
Qwen3 235B A22B (Non-reasoning)
14.0 coding index
Composite Indices
Higher is better; speed and price are normalized
Standard Benchmarks
Only benchmarks with data are shown
Differences That Matter
Best value
Qwen3 32B (Reasoning) has the strongest quality-to-price mix at 100.0 out of 100 value points.
Price gap
Qwen3 32B (Reasoning) is 2.3x cheaper on input tokens than Qwen3 235B A22B (Non-reasoning).
Speed gap
Qwen3 32B (Reasoning) generates about 1.2x as many tokens per second as Qwen3 235B A22B (Non-reasoning).
Reasoning gap
Qwen3 32B (Reasoning) leads Qwen3 235B A22B (Non-reasoning) by 23.0 points on reasoning.
Coding gap
Qwen3 235B A22B (Non-reasoning) leads Qwen3 32B (Reasoning) by 0.2 points on coding.
Response Face-Off
Run one prompt through the selected models and compare response quality with live speed and cost context.
Qwen3 235B A22B (Non-reasoning)
Alibaba
TTFT
—
Time
—
tok/s
—
Tokens
—
Cost
—
Qwen3 32B (Reasoning)
Alibaba
TTFT
—
Time
—
tok/s
—
Tokens
—
Cost
—
Which answer was more useful?
Full Comparison
| Metric | Al Qwen3 235B A22B (Non-reasoning) | Top Pick Al Qwen3 32B (Reasoning) |
|---|---|---|
| Pricing per 1M tokens | ||
| Input Cost | $0.45/1M | $0.20/1M |
| Output Cost | $1.80/1M | $0.52/1M |
| Blended (3:1) | $0.79/1M | $0.28/1M |
| Specifications | ||
| Organization | Alibaba | Alibaba |
| Release Date | Apr 28, 2025 | Apr 28, 2025 |
| Performance & Speed | ||
| Throughput | 66.3 tok/s | 78.8 tok/s |
| TTFT | 1241ms | 1096ms |
| Latency | 1241ms | 26470ms |
| Composite Indices | ||
| Value Score | 21.4 | 100.0 |
| Reasoning Score | 36.2 | 59.2 |
| Intelligence | 17.0 | 16.5 |
| Coding | 14.0 | 13.8 |
| Math | 23.7 | 73.0 |
| Standard Benchmarks | ||
| GPQA | 61.3% | 66.8% |
| MMLU Pro | 76.2% | 79.8% |
| HLE | 4.7% | 8.3% |
| LiveCodeBench | 34.3% | 54.6% |
| MATH 500 | 90.2% | 96.1% |
| AIME 2025 | 23.7% | 73.0% |
| AIME (Original) | 32.7% | 80.7% |
| SciCode | 29.9% | 35.4% |
| LCR | 0.0% | 0.0% |
| IFBench | 36.6% | 36.3% |
| TAU-bench v2 | 27.2% | 29.8% |
| TerminalBench Hard | 6.1% | 3.0% |
Key Takeaways
Qwen3 32B (Reasoning) offers the best value at $0.20/1M, making it ideal for high-volume applications and cost-conscious projects.
Qwen3 32B (Reasoning) has the strongest reasoning profile with a 59.2 reasoning score, combining the available reasoning-heavy benchmarks.
Qwen3 235B A22B (Non-reasoning) reaches a 14.0 coding index, making it the top choice for software development and code generation tasks.
All models support context windows of ∞+ tokens, suitable for processing lengthy documents and maintaining extended conversations.
When to Choose Each Model
Qwen3 235B A22B (Non-reasoning)
- Code generation
- Software development
Qwen3 32B (Reasoning)
- Cost-sensitive applications
- High-volume processing
- Complex reasoning tasks
- Research & analysis
