Model Comparison
GLM-4.7 (Non-reasoning)
vs. Qwen2 Instruct 72B
Comparing 2 AI models · 12 benchmarks · Z AI, Alibaba
Recommended Pick
Z GLM-4.7 (Non-reasoning) 8 metric wins Strongest on: Throughput, Reasoning, Intelligence
Lowest Price
Qwen2 Instruct 72B
$0.00/1M input price
Best Reasoning
GLM-4.7 (Non-reasoning)
40.5 reasoning score
Blends available reasoning benchmarks
Best for Coding
GLM-4.7 (Non-reasoning)
32.0 coding index
Composite Indices
Higher is better; speed and price are normalized
Standard Benchmarks
Only benchmarks with data are shown
Differences That Matter
Price gap
Qwen2 Instruct 72B is ∞x cheaper on input tokens than GLM-4.7 (Non-reasoning).
Reasoning gap
GLM-4.7 (Non-reasoning) leads Qwen2 Instruct 72B by 13.1 points on reasoning.
Top-pick rationale
GLM-4.7 (Non-reasoning) wins 8 measurable categories, including Throughput, Reasoning, Intelligence, GPQA.
Response Face-Off
Run one prompt through the selected models and compare response quality with live speed and cost context.
GLM-4.7 (Non-reasoning)
Z AI
TTFT
—
Time
—
tok/s
—
Tokens
—
Cost
—
Qwen2 Instruct 72B
Alibaba
TTFT
—
Time
—
tok/s
—
Tokens
—
Cost
—
Which answer was more useful?
Full Comparison
| Metric | Top Pick Z GLM-4.7 (Non-reasoning) | Al Qwen2 Instruct 72B |
|---|---|---|
| Pricing per 1M tokens | ||
| Input Cost | $0.60/1M | $0.00/1M |
| Output Cost | $2.20/1M | $0.00/1M |
| Blended (3:1) | $1.00/1M | — |
| Specifications | ||
| Organization | Z AI | Alibaba |
| Release Date | Dec 22, 2025 | Jun 7, 2024 |
| Performance & Speed | ||
| Throughput | 92.9 tok/s | — |
| TTFT | 828ms | — |
| Latency | 828ms | — |
| Composite Indices | ||
| Value Score | 100.0 | — |
| Reasoning Score | 40.5 | 27.5 |
| Intelligence | 34.2 | 11.7 |
| Coding | 32.0 | — |
| Math | 48.0 | — |
| Standard Benchmarks | ||
| GPQA | 66.4% | 37.1% |
| MMLU Pro | 79.4% | 62.2% |
| HLE | 6.1% | 3.7% |
| LiveCodeBench | 56.2% | 15.9% |
| MATH 500 | — | 70.1% |
| AIME 2025 | 48.0% | — |
| AIME (Original) | — | 14.7% |
| SciCode | 35.4% | 22.9% |
| LCR | 36.3% | — |
| IFBench | 54.6% | — |
| TAU-bench v2 | 94.2% | — |
| TerminalBench Hard | 30.3% | — |
Key Takeaways
Qwen2 Instruct 72B offers the best value at $0.00/1M, making it ideal for high-volume applications and cost-conscious projects.
GLM-4.7 (Non-reasoning) has the strongest reasoning profile with a 40.5 reasoning score, combining the available reasoning-heavy benchmarks.
GLM-4.7 (Non-reasoning) reaches a 32.0 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
GLM-4.7 (Non-reasoning)
- Complex reasoning tasks
- Research & analysis
- Code generation
- Software development
Qwen2 Instruct 72B
- Cost-sensitive applications
- High-volume processing
