Model Comparison
Qwen3 235B A22B 2507 Instruct
vs. Qwen3.5 122B A10B (Non-reasoning)
Comparing 2 AI models · 12 benchmarks · Alibaba
Recommended Pick
AlQwen3.5 122B A10B (Non-reasoning)12 metric winsStrongest on: Input price, Blended price, Throughput
Best Value
Qwen3 235B A22B 2507 Instruct
100.0 value score
59.6 reasoning / $1.22/1M
Lowest Price
Qwen3.5 122B A10B (Non-reasoning)
$0.40/1M input price
Best Reasoning
Qwen3 235B A22B 2507 Instruct
59.6 reasoning score
Blends available reasoning benchmarks
Best for Coding
Qwen3.5 122B A10B (Non-reasoning)
43.3 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 235B A22B 2507 Instruct has the strongest quality-to-price mix at 100.0 out of 100 value points.
Price gap
Qwen3.5 122B A10B (Non-reasoning) is 1.7x cheaper on input tokens than Qwen3 235B A22B 2507 Instruct.
Speed gap
Qwen3.5 122B A10B (Non-reasoning) generates about 2.5x as many tokens per second as Qwen3 235B A22B 2507 Instruct.
Reasoning gap
Qwen3 235B A22B 2507 Instruct leads Qwen3.5 122B A10B (Non-reasoning) by 17.9 points on reasoning.
Top-pick rationale
Qwen3.5 122B A10B (Non-reasoning) wins 12 measurable categories, including Input price, Blended price, Throughput, TTFT.
Response Face-Off
Run one prompt through the selected models and compare response quality with live speed and cost context.
Qwen3 235B A22B 2507 Instruct
Alibaba
TTFT
—
Time
—
tok/s
—
Tokens
—
Cost
—
Qwen3.5 122B A10B (Non-reasoning)
Alibaba
TTFT
—
Time
—
tok/s
—
Tokens
—
Cost
—
Which answer was more useful?
Chat with leading AI models
Use Claude, ChatGPT, Gemini alongside with EU-Hosted Models like Deepseek, Qwen & Kimi.
EU-hosted inference
Servers in Germany & Finland. Designed to meet strict GDPR and ISO 27001 compliance requirements.
Full Comparison
| Metric | Al Qwen3 235B A22B 2507 Instruct | Top Pick Al Qwen3.5 122B A10B (Non-reasoning) |
|---|---|---|
| Pricing per 1M tokens | ||
| Input Cost | $0.70/1M | $0.40/1M |
| Output Cost | $2.80/1M | $3.20/1M |
| Blended (3:1) | $1.22/1M | $1.10/1M |
| Specifications | ||
| Organization | Alibaba | Alibaba |
| Release Date | Jul 21, 2025 | Feb 24, 2026 |
| Performance & Speed | ||
| Throughput | 65.2 tok/s | 165.8 tok/s |
| TTFT | 1154ms | 1024ms |
| Latency | 1154ms | 1024ms |
| Composite Indices | ||
| Value Score | 100.0 | 77.9 |
| Reasoning Score | 59.6 | 41.7 |
| Intelligence | 18.2 | 27.6 |
| Coding | — | 43.3 |
| Math | 71.7 | — |
| Standard Benchmarks | ||
| GPQA | 75.3% | 82.7% |
| MMLU Pro | 82.8% | — |
| HLE | 10.6% | 14.8% |
| LiveCodeBench | 52.4% | — |
| MATH 500 | 98.0% | — |
| AIME 2025 | 71.7% | — |
| AIME (Original) | 71.7% | — |
| SciCode | 36.0% | 35.6% |
| LCR | 31.2% | 56.0% |
| IFBench | 46.1% | 50.8% |
| TAU-bench v2 | 33.3% | 84.5% |
| TerminalBench Hard | 15.2% | 29.5% |
Key Takeaways
Qwen3.5 122B A10B (Non-reasoning) offers the best value at $0.40/1M,making it ideal for high-volume applications and cost-conscious projects.
Qwen3 235B A22B 2507 Instruct has the strongest reasoning profile with a 59.6 reasoning score,combining the available reasoning-heavy benchmarks.
Qwen3.5 122B A10B (Non-reasoning) reaches a 43.3 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 2507 Instruct
- Complex reasoning tasks
- Research & analysis
Qwen3.5 122B A10B (Non-reasoning)
- Cost-sensitive applications
- High-volume processing
- Code generation
- Software development
