Model Comparison
gpt-oss-120b (low)
vs. Qwen3.6 35B A3B (Non-reasoning)
Comparing 2 AI models · 10 benchmarks · OpenAI, Alibaba
Recommended Pick
Op gpt-oss-120b (low) 9 metric wins Strongest on: Value, Input price, Output price
Best Value
gpt-oss-120b (low)
100.0 value score
46.1 reasoning / $0.26/1M
Lowest Price
gpt-oss-120b (low)
$0.15/1M input price
Best Reasoning
gpt-oss-120b (low)
46.1 reasoning score
Blends available reasoning benchmarks
Best for Coding
Qwen3.6 35B A3B (Non-reasoning)
17.6 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
gpt-oss-120b (low) has the strongest quality-to-price mix at 100.0 out of 100 value points.
Price gap
gpt-oss-120b (low) is 2.5x cheaper on input tokens than Qwen3.6 35B A3B (Non-reasoning).
Speed gap
gpt-oss-120b (low) generates about 2.2x as many tokens per second as Qwen3.6 35B A3B (Non-reasoning).
Reasoning gap
gpt-oss-120b (low) leads Qwen3.6 35B A3B (Non-reasoning) by 4.2 points on reasoning.
Coding gap
Qwen3.6 35B A3B (Non-reasoning) leads gpt-oss-120b (low) by 2.1 points on coding.
Response Face-Off
Run one prompt through the selected models and compare response quality with live speed and cost context.
gpt-oss-120b (low)
OpenAI
TTFT
—
Time
—
tok/s
—
Tokens
—
Cost
—
Qwen3.6 35B A3B (Non-reasoning)
Alibaba
TTFT
—
Time
—
tok/s
—
Tokens
—
Cost
—
Which answer was more useful?
Full Comparison
| Metric | Top Pick Op gpt-oss-120b (low) | Al Qwen3.6 35B A3B (Non-reasoning) |
|---|---|---|
| Pricing per 1M tokens | ||
| Input Cost | $0.15/1M | $0.38/1M |
| Output Cost | $0.60/1M | $2.25/1M |
| Blended (3:1) | $0.26/1M | $0.84/1M |
| Specifications | ||
| Organization | OpenAI | Alibaba |
| Release Date | Aug 5, 2025 | Apr 16, 2026 |
| Performance & Speed | ||
| Throughput | 345.2 tok/s | 153.8 tok/s |
| TTFT | 479ms | 1414ms |
| Latency | 6273ms | 1414ms |
| Composite Indices | ||
| Value Score | 100.0 | 28.3 |
| Reasoning Score | 46.1 | 41.9 |
| Intelligence | 24.5 | 31.5 |
| Coding | 15.5 | 17.6 |
| Math | 66.7 | — |
| Standard Benchmarks | ||
| GPQA | 67.2% | 81.7% |
| MMLU Pro | 77.5% | — |
| HLE | 5.2% | 12.5% |
| LiveCodeBench | 70.7% | — |
| AIME 2025 | 66.7% | — |
| SciCode | 36.0% | 1.3% |
| LCR | 43.7% | 56.7% |
| IFBench | 58.3% | 36.2% |
| TAU-bench v2 | 45.0% | 85.1% |
| TerminalBench Hard | 5.3% | 25.8% |
Key Takeaways
gpt-oss-120b (low) offers the best value at $0.15/1M, making it ideal for high-volume applications and cost-conscious projects.
gpt-oss-120b (low) has the strongest reasoning profile with a 46.1 reasoning score, combining the available reasoning-heavy benchmarks.
Qwen3.6 35B A3B (Non-reasoning) reaches a 17.6 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
gpt-oss-120b (low)
- Cost-sensitive applications
- High-volume processing
- Complex reasoning tasks
- Research & analysis
Qwen3.6 35B A3B (Non-reasoning)
- Code generation
- Software development
